(19)
(11)EP 3 252 767 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
26.06.2019 Bulletin 2019/26

(21)Application number: 17171205.2

(22)Date of filing:  16.05.2017
(51)International Patent Classification (IPC): 
G10L 21/038(2013.01)
G10L 19/24(2013.01)
G10L 19/16(2013.01)

(54)

VOICE SIGNAL PROCESSING METHOD, RELATED APPARATUS, AND SYSTEM

SPRACHSIGNALVERARBEITUNGSVERFAHREN, ZUGEHÖRIGE VORRICHTUNG UND SYSTEM

PROCÉDÉ DE TRAITEMENT DE SIGNAL VOCAL, APPAREIL ASSOCIÉ, ET SYSTÈME


(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30)Priority: 31.05.2016 CN 201610379386

(43)Date of publication of application:
06.12.2017 Bulletin 2017/49

(73)Proprietor: Huawei Technologies Co., Ltd.
Longgang District Shenzhen, Guangdong 518129 (CN)

(72)Inventors:
  • WANG, Bin
    Shenzhen, Guangdong 518129 (CN)
  • XIA, Bingyin
    Shenzhen, Guangdong 518129 (CN)
  • LIU, Zexin
    Shenzhen, Guangdong 518129 (CN)
  • MIAO, Lei
    Shenzhen, Guangdong 518129 (CN)

(74)Representative: Goddar, Heinz J. 
Boehmert & Boehmert Anwaltspartnerschaft mbB Pettenkoferstrasse 22
80336 München
80336 München (DE)


(56)References cited: : 
US-A1- 2004 254 786
US-B1- 7 461 003
US-A1- 2005 267 739
  
  • LAAKSONEN L ET AL: "Development, evaluation and implementation of an artificial bandwidth extension method of telephone speech in mobile terminal", IEEE TRANSACTIONS ON CONSUMER ELECTRONICS, IEEE SERVICE CENTER, NEW YORK, NY, US, vol. 52, no. 2, 1 May 2009 (2009-05-01), pages 780-787, XP011270938, ISSN: 0098-3063
  • JAX P ET AL: "Bandwidth Extension of Speech Signals: A Catalyst for the Introduction of Wideband Speech Coding?", IEEE COMMUNICATIONS MAGAZINE, IEEE SERVICE CENTER, PISCATAWAY, US, vol. 44, no. 5, 1 May 2006 (2006-05-01), pages 106-111, XP001546248, ISSN: 0163-6804, DOI: 10.1109/MCOM.2006.1637954
  • VASEGHI S ET AL: "Speech Bandwidth Extension: Extrapolations of Spectral Envelop and Harmonicity Quality of Excitation", ACOUSTICS, SPEECH AND SIGNAL PROCESSING, 2006. ICASSP 2006 PROCEEDINGS . 2006 IEEE INTERNATIONAL CONFERENCE ON TOULOUSE, FRANCE 14-19 MAY 2006, PISCATAWAY, NJ, USA,IEEE, PISCATAWAY, NJ, USA, 14 May 2006 (2006-05-14), page III, XP031386788, ISBN: 978-1-4244-0469-8
  
Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


Description

TECHNICAL FIELD



[0001] The present invention relates to the field of audio technologies, and in particular, to a voice signal processing method, a related apparatus, and a system.

BACKGROUND



[0002] Because of limitation of a transmission bandwidth and other conditions, currently, a frequency bandwidth of a voice signal transmitted on a main telephony network is generally less than 4 kHz, and a frequency band is generally limited to a range of 300 Hz to 3.4 kHz. As a communication bandwidth gradually increases, users impose an increasingly high requirement for voice quality and immersive experience. Consequently, a conventional narrowband voice can no longer meet an experience requirement of a user. In addition, lack of high frequency information in a conventional telephony voice greatly affects users with hearing impairments, and it is usually difficult for the users with hearing impairments to talk over a telephone. Therefore, a wideband voice, even a super wideband voice, and the like become increasingly popular.

[0003] Nowadays, 2nd generation (2G), 3rd generation (3G), and 4th generation (4G) networks coexist, and therefore terminals that support various voice bandwidths may coexist. For example, a narrowband terminal with a maximum support capability for a narrowband (NB, Narrowband) voice bandwidth, a wideband terminal with a maximum support capability for a wideband (WB, Wideband) voice bandwidth, a super wideband terminal with a maximum support capability for a super wideband (SWB, Super Wideband) voice bandwidth, and a full band terminal with a maximum support capability for a full band (FB, Full Band) voice bandwidth may coexist.

[0004] When a call is placed between two terminals that support a same maximum voice bandwidth, a corresponding voice signal bandwidth service may be established. However, when a call is placed between a terminal supporting a relatively narrow maximum voice bandwidth (for example, NB) and a terminal supporting a relatively wide maximum voice bandwidth (such as WB or SWB), in the conventional solution, the terminal (for example, a WB terminal) supporting a relatively wide maximum voice bandwidth usually can enjoy a voice bandwidth service that is only basically equal to a voice bandwidth service (for example, an NB bandwidth service) of the terminal (for example, an NB terminal) supporting a relatively narrow maximum voice bandwidth.

[0005] US 2004/254786 A1 discloses a method for transcoding audio signals in a communications system. In order to improve the inter-operability between units capable of handling wideband audio signals and units or network components capable of handling narrowband audio signals, it is proposed that first, an audio signal is received in a network element of a communications network via which said audio signal is transmitted. Next, it is determined in said network element whether a transcoding of the received audio signal is required. In case a narrowband-to-wideband transcoding of the received signal is required, the received narrowband audio signal is transcoded into a wideband audio signal in the network element. The generated wideband audio signal is then forwarded to the receiving terminal. The disclosure equally relates to a corresponding communications system and its components.

[0006] US 7461 003 B1 discloses methods and apparatus to extend the bandwidth of a speech communication to yield a perceived higher quality speech communication for an enhanced user experience. In one aspect of the invention, for example, methods and apparatus can be used to extend the bandwidth of a speech communication beyond a band-limited region defined by the lowest limit and highest limit of the frequency spectrum by which such speech communication is otherwise characterized absent such bandwidth extension. In another aspect of the invention, for example, methods and apparatus can be used to substitute for corrupt, missing or lost components of a given speech communication, or to otherwise enhance the perceived quality of a speech communication, by extending the speech communication to include one or more artificially created points within the region defined by the lowest limit and highest limit of the frequency spectrum by which such speech communication is characterized. The result is a speech communication that is perceived to be of higher quality. The various aspects of the present invention can be applied, for example, to network devices or to end-terminal devices.

[0007] LAAKSONEN L ET AL: "Development, evaluation and implementation of an artificial bandwidth extension method of telephone speech in mobile terminal", IEEE TRANSACTIONS ON CONSUMER ELECTRONICS, IEEE SERVICE CENTER, NEW YORK, NY, US, vol. 52, no. 2,1 May 2009 (2009-05-01), pages 780-787, ISSN: 0098-3063 discloses that artificial bandwidth extension methods aim to improve the quality and intelligibility of narrowband telephone speech by adding new, artificially generated spectral content to the highband of the received voice signal. The development cycle of an artificial bandwidth extension method from the initial idea to the implementation in a mobile terminal is discussed in this paper. Developing the algorithm in the Matlab environment is the first step in the process. The method was then evaluated in formal listening tests and simulations to verify its performance in different scenarios. Finally, the utilization of this technology in a product included its DSP implementation combined with the acoustical design of the user terminal.

[0008] JAX P ET AL: "Bandwidth Extension of Introduction of Wideband Speech Coding?", IEEE COMMUNICATIONS MAGAZINE, IEEE SERVICE CENTER, PISCATAWAY, US, vol. 44, no. 5,1 May 2006 (2006-05-01), pages 106-111, ISSN: 0163-6804, DOI: 10.1109/MCOM.2006.1637954 discloses basic principles of bandwidth extension, and discuss several application scenarios in which both wideband coding and BWE complement each other.

[0009] US 2005/0267739 A1 discloses artificial bandwidth expansion devices, Systems, methods and computer code products for expanding a narrowband speech signal into an artificially expanded wide- band speech signal. Embodiments of the invention can operate by forming an unshaped wideband signal based on the narrowband speech signal, such as through aliasing, and shaping the wideband signal into the artificially expanded wideband speech signal by amplifying/attenuating the unshaped wideband signal using a function generated by a neural network. Weights of the neural network can be set by a training/learning subsystem which generates genomes containing the neural network weights based on simulated environments in which a device employing the artificial bandwidth expansion is expected to operate.

SUMMARY



[0010] Embodiments of the present invention provide a voice signal processing method, a related apparatus, and a system.

[0011] A first aspect of the embodiments of the present invention provides a voice signal processing method, including:

receiving, by a network device, a first voice coded signal from a first terminal; performing, by the network device, voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal; performing, by the network device by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal; after combining the first voice decoded signal and the bandwidth extension voice decoded signal, performing, by the network device, voice coding processing to obtain a second voice coded signal; and sending, by the network device, the second voice coded signal to a second terminal that establishes a call connection to the first terminal, where a maximum frequency bandwidth supported by the first terminal is less than a maximum frequency bandwidth supported by the second terminal,

wherein the performing (203, 304, 404), by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal comprises estimating, by using the voice decoding parameter, a bandwidth extension excitation signal corresponding to the first voice decoded signal, wherein the estimating the bandwidth extension excitation signal comprises estimating, by using the voice decoding parameter and based on a harmonic noise model algorithm, the bandwidth extension excitation signal corresponding to the first voice decoded signal; estimating, by using the voice decoding parameter, a bandwidth extension spectral envelope corresponding to the first voice decoded signal; and performing, by using a filter corresponding to the bandwidth extension spectral envelope, synthesis processing on the bandwidth extension excitation signal to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal; and

selecting, from multiple virtual bandwidth extension processors, a virtual bandwidth extension processor corresponding to the maximum frequency bandwidth supported by the second terminal, and performing, by using the voice decoding parameter, the virtual bandwidth extension processing using the selected virtual bandwidth extension processor to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal.



[0012] For example, the maximum frequency bandwidth supported by the first terminal and the maximum frequency bandwidth supported by the second terminal may be two of the following typical frequency bandwidths: narrowband (NB), wideband (WB), super wideband (SWB), or full band (FB). That is, the first terminal may be, a narrowband terminal, a wideband terminal, or a super wideband terminal, and the second terminal may be a wideband terminal, a super wideband terminal, or a full band terminal. Certainly, the maximum frequency bandwidth supported by the first terminal and the maximum frequency bandwidth supported by the second terminal are not limited to the typical frequency bandwidths in the foregoing example.

[0013] A frequency bandwidth of the first voice coded signal is less than a frequency bandwidth of the second voice coded signal, and a sampling rate of the first voice coded signal is less than a sampling rate of the second voice coded signal.

[0014] The frequency bandwidth of the first voice coded signal is less than or equal to the maximum frequency bandwidth supported by the first terminal. For example, when the first terminal is a narrowband terminal, the first voice coded signal may be a narrowband voice coded signal. When the first terminal is a wideband terminal, the first voice coded signal may be a wideband voice coded signal or a narrowband voice coded signal. When the first terminal is a super wideband terminal, the first voice coded signal may be a super wideband voice coded signal, a narrowband voice coded signal, or a wideband voice coded signal.

[0015] The frequency bandwidth of the first voice coded signal is less than the frequency bandwidth of the second voice coded signal. The sampling rate of the first voice coded signal is less than the sampling rate of the second voice coded signal. For example, a frequency band of the first voice coded signal may be a subset of a frequency band of the second voice coded signal. Certainly, an intersection set between the frequency band of the first voice coded signal and the frequency band of the second voice coded signal may not be equal to the frequency band of the first voice coded signal.

[0016] The bandwidth extension voice decoded signal may include a high bandwidth extension voice decoded signal, and the bandwidth extension voice decoded signal may further include a low bandwidth extension voice decoded signal. For example, it is assumed that a frequency bandwidth of the first voice decoded signal is 3400 Hz-300 Hz=3100 Hz. A bandwidth extension voice decoded signal corresponding to the first voice decoded signal may include the high bandwidth extension voice decoded signal (for example, 7000 Hz-3400 Hz=3600 Hz), and a bandwidth extension voice decoded signal corresponding to the first voice decoded signal may further include the low bandwidth extension voice decoded signal (for example, 300 Hz-50 Hz=250 Hz). For another example, it is assumed that a frequency bandwidth of the first voice decoded signal is 7000 Hz-50 Hz=6950 Hz. A bandwidth extension voice decoded signal corresponding to the first voice decoded signal may include the high bandwidth extension voice decoded signal (specifically, for example, 14000 Hz-7000 Hz=7000 Hz). For another example, it is assumed that a frequency bandwidth of the first voice decoded signal is 14 kHz-50 Hz=13950 Hz. A bandwidth extension voice decoded signal corresponding to the first voice decoded signal may include the high bandwidth extension voice decoded signal (specifically, for example, 24 kHz-14 kHz=10 kHz). Another situation may be learned by analogy.

[0017] The network device involved in this embodiment may be, for example, a base station, a radio network controller, a core network device, or another network device. For example, the network device may be specifically a base station or a radio network controller in a radio access network accessed by the second terminal, or a base station or a radio network controller in a radio access network accessed by the first terminal, or may be a core network device, such as a packet data gateway or a serving gateway.

[0018] A first terminal and a second terminal may be user equipment that has a call function, such as a mobile phone, a tablet computer, a personal computer, or a notebook computer.

[0019] It can be learned that, in the foregoing technical solution, after receiving a first voice coded signal from a first terminal supporting a relatively narrow bandwidth, a network device performs voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal. The network device performs, by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal. Then, after combining the first voice decoded signal and the bandwidth extension voice decoded signal, the network device performs voice coding processing to obtain a second voice coded signal, and then sends the second voice coded signal to a second terminal supporting a relatively wide bandwidth. The network device at a transit location performs virtual bandwidth extension on a voice coded signal that is sent by the first terminal supporting a relatively narrow bandwidth to the second terminal supporting a relatively wide bandwidth. Therefore, further, a downlink voice coded signal of the second terminal supporting a relatively wide bandwidth can better match a maximum frequency bandwidth support capability of the second terminal, so that the second terminal supporting a relatively wide bandwidth can enjoy a voice signal bandwidth service that matches the maximum frequency bandwidth support capability of the second terminal. In addition, special function enhancement does not need to be performed on the second terminal, thereby improving call experience of a user. It can be learned that, in the foregoing example, service quality of terminals that have asymmetric maximum frequency bandwidth support capabilities can be improved;

[0020] In a first possible implementation manner of the first aspect, the estimating, by using the voice decoding parameter, a bandwidth extension spectral envelope corresponding to the first voice decoded signal includes: estimating, by using the voice decoding parameter and based on a linear mapping method, a codebook mapping method, or a statistics mapping method, the bandwidth extension spectral envelope corresponding to the first voice decoded signal.

[0021] In a second possible implementation manner of the first aspect, the voice decoding parameter includes a pitch period, a voicing factor, and a linear predictive coding parameter.

[0022] In a third possible implementation manner of the first aspect, the performing voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal includes: selecting, from multiple voice decoders, a voice decoder corresponding to the maximum frequency bandwidth supported by the first terminal, and performing the voice decoding processing on the first voice coded signal to obtain the voice decoding parameter and the first voice decoded signal.

[0023] It can be understood that multiple voice decoders are built in a network device; and selecting, according to a requirement, an appropriate voice decoder to perform decoding on a voice coded signal helps improve a transcoding support capability and a response processing speed of the network device.

[0024] In a further possible implementation manner of the first aspect, after combining the first voice decoded signal and the bandwidth extension voice decoded signal, the performing voice coding processing to obtain a second voice coded signal includes: selecting, from multiple voice encoders, a voice encoder corresponding to the maximum frequency bandwidth supported by the second terminal, and after combining the first voice decoded signal and the bandwidth extension voice decoded signal, performing the voice coding processing to obtain the second voice coded signal.

[0025] In a further possible implementation manner of the first aspect, frequency bandwidths of the first voice coded signal and the second voice coded signal are two of the following frequency bandwidths: narrowband, wideband, super wideband, or full band.

[0026] In a further possible implementation manner of the first aspect, the voice signal processing method further includes:

receiving, by the network device, a third voice coded signal from the second terminal;

performing, by the network device, voice decoding processing on the third voice coded signal to obtain a third voice decoded signal;

performing, by the network device, down-sampling processing on the third voice decoded signal to obtain a fourth voice decoded signal; and performing, by the network device, voice coding processing on the fourth voice decoded signal to obtain a fourth voice coded signal, where a frequency bandwidth of the fourth voice coded signal is less than a frequency bandwidth of the third voice coded signal, and a sampling rate of the fourth voice coded signal is less than a sampling rate of the third voice coded signal; and

sending, by the network device, the fourth voice coded signal to the first terminal; or after the network device performs voice enhancement processing on the fourth voice coded signal to obtain a fourth voice coded signal obtained after the voice enhancement processing, sending the fourth voice coded signal obtained after the voice enhancement processing to the first terminal.



[0027] It can be understood that performing voice enhancement processing on a voice coded signal sent to a first terminal helps improve a gain of a voice coded signal received by the first terminal, and further helps improve call experience of a terminal with a support capability for a relatively narrow bandwidth.

[0028] In a further possible implementation manner of the first aspect, the performing voice decoding processing on the third voice coded signal to obtain a third voice decoded signal includes: selecting, from multiple voice decoders, a voice decoder corresponding to the maximum frequency bandwidth supported by the second terminal, and performing the voice decoding processing on the third voice coded signal to obtain the third voice decoded signal; or the performing down-sampling processing on the third voice decoded signal to obtain a fourth voice decoded signal includes: selecting, from multiple down-samplers, a down-sampler corresponding to the maximum frequency bandwidth supported by the first terminal, and performing the down-sampling processing on the third voice decoded signal to obtain the fourth voice decoded signal; or
the performing voice coding processing on the fourth voice decoded signal to obtain a fourth voice coded signal includes: selecting, from multiple voice encoders, a voice encoder corresponding to the maximum frequency bandwidth supported by the first terminal, and performing the voice coding processing on the fourth voice decoded signal to obtain the fourth voice coded signal.

[0029] In a further possible implementation manner of the first aspect, the network device is a base station, a radio network controller, or a core network device.

[0030] A second aspect of the embodiments of the present invention provides a network device, including:

a communications interface, configured to receive a first voice coded signal from a first terminal;

a first voice decoder, configured to perform voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal;

a first virtual bandwidth extension processor, configured to perform, by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal; and

a first voice encoder, configured to: after combining the first voice decoded signal and the bandwidth extension voice decoded signal, perform voice coding processing to obtain a second voice coded signal, where a frequency bandwidth of the first voice coded signal is less than a frequency bandwidth of the second voice coded signal, and a sampling rate of the first voice coded signal is less than a sampling rate of the second voice coded signal; where

the communications interface is further configured to send the second voice coded signal to a second terminal that establishes a call connection to the first terminal, and a maximum frequency bandwidth supported by the first terminal is less than a maximum frequency bandwidth supported by the second terminal,

wherein the first virtual bandwidth extension processor (530) is specifically configured to:

estimate, by using the voice decoding parameter, a bandwidth extension excitation signal corresponding to the first voice decoded signal, wherein

in a process of estimating, by using the voice decoding parameter, the bandwidth extension excitation signal corresponding to the first voice decoded signal, the first virtual bandwidth extension processor (530) is specifically configured to estimate, by using the voice decoding parameter and based on a harmonic noise model algorithm, the bandwidth extension excitation signal corresponding to the first voice decoded signal;

estimate, by using the voice decoding parameter, a bandwidth extension spectral envelope corresponding to the first voice decoded signal; and perform, by using a filter corresponding to the bandwidth extension spectral envelope, synthesis processing on the bandwidth extension excitation signal to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal;

wherein the network device is further configured to select, from multiple virtual bandwidth extension processors, the first virtual bandwidth extension processor as the virtual bandwidth extension processor corresponding to the maximum frequency bandwidth supported by the second terminal, and perform, by using the voice decoding parameter, the virtual bandwidth extension processing to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal



[0031] It can be understood that a first voice decoder, a first virtual bandwidth extension processor, and a first voice encoder may be integrated.

[0032] In a first possible implementation manner of the second aspect,
in a process of estimating, by using the voice decoding parameter, the bandwidth extension spectral envelope corresponding to the first voice decoded signal, the first virtual bandwidth extension processor is specifically configured to estimate, by using the voice decoding parameter and based on a linear mapping method, a codebook mapping method, or a statistics mapping method, the bandwidth extension spectral envelope corresponding to the first voice decoded signal.

[0033] In a second possible implementation manner of the second aspect, the voice decoding parameter includes a pitch period, a voicing factor, and a linear predictive coding parameter.

[0034] In a third possible implementation manner of the second aspect, the network device includes multiple voice decoders, and the first voice decoder is a voice decoder that is in multiple voice decoders and that is corresponding to the maximum frequency bandwidth supported by the first terminal; or,
the network device includes multiple virtual bandwidth extension processors, and the first virtual bandwidth extension processor is a virtual bandwidth extension processor that is in multiple virtual bandwidth extension processors and that is corresponding to the maximum frequency bandwidth supported by the second terminal; or
the network device includes multiple voice encoders, and the first voice encoder is a voice encoder that is in multiple voice encoders and that is corresponding to the maximum frequency bandwidth supported by the second terminal.

[0035] In a fourth possible implementation manner of the second aspect, frequency bandwidths of the first voice coded signal and the second voice coded signal are two of the following frequency bandwidths: narrowband, wideband, super wideband, or full band.

[0036] In a fifth possible implementation manner of the second aspect,
the network device further includes a second voice decoder, a second voice encoder, and a first down-sampler;
the communications interface is further configured to receive a third voice coded signal from the second terminal;
the second voice decoder is configured to perform voice decoding processing on the third voice coded signal to obtain a third voice decoded signal;
the first down-sampler is configured to perform down-sampling processing on the third voice decoded signal to obtain a fourth voice decoded signal;
the second voice encoder is configured to perform voice coding processing on the fourth voice decoded signal to obtain a fourth voice coded signal, where a frequency bandwidth of the fourth voice coded signal is less than a frequency bandwidth of the third voice coded signal, and a sampling rate of the fourth voice coded signal is less than a sampling rate of the third voice coded signal; and
the communications interface is further configured to send the fourth voice coded signal to the first terminal; or the communications interface is further configured to: after performing voice enhancement processing on the fourth voice coded signal to obtain a fourth voice coded signal obtained after the voice enhancement processing, send the fourth voice coded signal obtained after the voice enhancement processing to the first terminal.

[0037] In a sixth possible implementation manner of the second aspect, the network device includes multiple voice decoders, and the second voice decoder is a voice decoder corresponding to the maximum frequency bandwidth supported by the second terminal; or
the network device includes multiple down-samplers, and the first down-sampler is a down-sampler that is in multiple down-samplers and that is corresponding to the maximum frequency bandwidth supported by the first terminal; or
the network device includes multiple voice encoders, and the second voice encoder is a voice encoder corresponding to the maximum frequency bandwidth supported by the first terminal.

[0038] In a seventh possible implementation manner of the second aspect, the network device is a base station, a radio network controller, or a core network device.

[0039] A sixth aspect of the embodiments of the present invention provides a communications system, including any network device provided in the embodiments of the present invention.

BRIEF DESCRIPTION OF DRAWINGS



[0040] To describe the technical solutions in the embodiments of the present invention more clearly, the following briefly describes the accompanying drawings required for describing the embodiments. Apparently, the accompanying drawings in the following description show merely some embodiments of the present invention, and a person of ordinary skill in the art may still derive other drawings from these accompanying drawings without creative efforts.

FIG. 1-A and FIG. 1-B are schematic diagrams of two network architectures according to an embodiment of the present invention;

FIG. 2-A is a schematic diagram of a voice signal processing method according to an embodiment of the present invention;

FIG. 2-B is a schematic diagram of an example of a spectrum range of a typical frequency band according to an embodiment of the present invention;

FIG. 3-A is a schematic diagram of another voice signal processing method according to an embodiment of the present invention;

FIG. 3-B is a schematic diagram illustrating internal devices of a network device according to an embodiment of the present invention;

FIG. 3-C is a schematic diagram of a spectrum range of narrowband according to an embodiment of the present invention;

FIG. 3-D is a schematic diagram of a spectrum range of wideband extended from narrowband according to an embodiment of the present invention;

FIG. 3-E is a schematic diagram of a flow direction of a voice signal according to an embodiment of the present invention;

FIG. 4-A is a schematic diagram of another voice signal processing method according to an embodiment of the present invention;

FIG. 4-B is a schematic diagram of a flow direction of a voice signal according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a network device according to an embodiment of the present invention; and

FIG. 6 is a schematic structural diagram of another network device according to an embodiment of the present invention.


DESCRIPTION OF EMBODIMENTS



[0041] The terms "include" and "have" and any variants thereof in the specification, claims, and accompanying drawings of the present invention are intended to cover non-exclusive inclusion. For example, a process, a method, a system, a product, or a device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes an unlisted step or unit, or optionally further includes another inherent step or unit of the process, the method, the product, or the device. In addition, the terms "first", "second", "third", and so on are intended to distinguish between different objects but are not intended to describe a specific order.

[0042] In the following, first, referring to FIG. 1-A and FIG. 1-B, FIG. 1-A and FIG. 1-Bare schematic diagrams of two possible network architectures according to an embodiment of this application. In the network architecture shown in FIG. 1-A, a call connection is established between terminals by using an access network and a core network. The voice signal processing methods provided in some embodiments of this application may be performed by an access network device or a core network device. For example, the voice signal processing methods provided in some embodiments of this application may be performed by some servers (such as a conferencing server or a network telephony server) on the Internet.

[0043] The terminal involved in this embodiment of the present invention may be an apparatus that has a function such as collecting, storing, or transmitting a voice signal. Specifically, the terminal may be, for example, a mobile phone, a tablet computer, a personal computer, or a notebook computer.

[0044] Referring to FIG. 2-A, FIG. 2-A is a schematic flowchart of a voice signal processing method according to an embodiment of this application. As shown in an example in FIG. 2-A, the voice signal processing method provided in this embodiment of this application may include the following steps.

[0045] 201. A first terminal sends a first voice coded signal.

[0046] After a call connection is established between the first terminal and a second terminal, the first terminal may send the first voice coded signal based on the call connection. The second terminal may also send a voice coded signal based on the call connection.

[0047] The call connection may be a call connection based on a mobile communications network, or may be a call connection based on the Internet.

[0048] In the example of this embodiment of this application, a maximum frequency bandwidth supported by the first terminal is less than a maximum frequency bandwidth supported by the second terminal.

[0049] For example, the maximum frequency bandwidth supported by the first terminal and the maximum frequency bandwidth supported by the second terminal may be two of the following typical frequency bandwidths: narrowband (NB), wideband (WB), super wideband (SWB), or full band (FB). That is, the first terminal may be, a narrowband terminal, a wideband terminal, or a super wideband terminal, and the second terminal may be a wideband terminal, a super wideband terminal, or a full band terminal. Certainly, the maximum frequency bandwidth supported by the first terminal and the maximum frequency bandwidth supported by the second terminal are not limited to the typical frequency bandwidths in the foregoing example.

[0050] Referring to FIG. 2-B, FIG. 2-B shows an example of a range of a typical frequency bandwidth, such as narrowband (NB), wideband (WB), super wideband (SWB), and full band (FB). Certainly, a range boundary of these typical frequency bands is not limited to an example shown in the figure.

[0051] A frequency bandwidth of the first voice coded signal is less than or equal to the maximum frequency bandwidth supported by the first terminal. For example, when the first terminal is a narrowband terminal, the first voice coded signal may be a narrowband voice coded signal. When the first terminal is a wideband terminal, the first voice coded signal may be a wideband voice coded signal or a narrowband voice coded signal. When the first terminal is a super wideband terminal, the first voice coded signal may be a super wideband voice coded signal, a narrowband voice coded signal, or a wideband voice coded signal.

[0052] 202. The network device receives the first voice coded signal from the first terminal, and the network device performs voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal.

[0053] For example, the voice decoding parameter may include a pitch period, a voicing factor, and a linear predictive coding parameter.

[0054] 203. The network device performs, by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal.

[0055] The virtual bandwidth extension processing is performed by using the voice decoding parameter to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal, and a main mechanism is use of correlation between a low frequency band voice signal and a high frequency band voice signal. Therefore, various related bandwidth extension algorithms based on the foregoing mechanism may be optionally used to perform the VBWE (Virtual Bandwidth Extension, virtual bandwidth extension) processing.

[0056] The bandwidth extension voice decoded signal may include a high bandwidth extension voice decoded signal, and the bandwidth extension voice decoded signal may further include a low bandwidth extension voice decoded signal. For example, it is assumed that a frequency bandwidth of the first voice decoded signal is 3400 Hz-300 Hz=3100 Hz. A bandwidth extension voice decoded signal corresponding to the first voice decoded signal may include the high bandwidth extension voice decoded signal (for example, 7000 Hz-3400 Hz=3600 Hz), and a bandwidth extension voice decoded signal corresponding to the first voice decoded signal may further include the low bandwidth extension voice decoded signal (for example, 300 Hz-50 Hz=250 Hz). For another example, it is assumed that a frequency bandwidth of the first voice decoded signal is 7000 Hz-50 Hz=6950 Hz. A bandwidth extension voice decoded signal corresponding to the first voice decoded signal may include the high bandwidth extension voice decoded signal (specifically, for example, 14000 Hz-7000 Hz=7000 Hz). For another example, it is assumed that a frequency bandwidth of the first voice decoded signal is 14 kHz-50 Hz=13950 Hz. A bandwidth extension voice decoded signal corresponding to the first voice decoded signal may include the high bandwidth extension voice decoded signal (specifically, for example, 24 kHz-14 kHz=10 kHz). Another situation may be learned by analogy.

[0057] 204. After combining the first voice decoded signal and the bandwidth extension voice decoded signal, the network device performs voice coding processing to obtain a second voice coded signal.

[0058] The frequency bandwidth of the first voice coded signal is less than a frequency bandwidth of the second voice coded signal. A sampling rate of the first voice coded signal is less than a sampling rate of the second voice coded signal. For example, a frequency band of the first voice coded signal may be a subset of a frequency band of the second voice coded signal. Certainly, an intersection set between the frequency band of the first voice coded signal and the frequency band of the second voice coded signal may not be equal to the frequency band of the first voice coded signal.

[0059] 205. The network device sends the second voice coded signal to a second terminal that establishes a call connection to the first terminal.

[0060] Correspondingly, the second terminal may receive the second voice coded signal, and perform decoding on the second voice coded signal and play a signal obtained after the decoding.

[0061] The network device involved in this embodiment may be, for example, a base station, a radio network controller, a core network device, or another network device. For example, the network device may be specifically a base station or a radio network controller in a radio access network accessed by the second terminal, or a base station or a radio network controller in a radio access network accessed by the first terminal, or may be a core network device, such as a packet data gateway or a serving gateway.

[0062] A first terminal and a second terminal may be user equipment that has a call function, such as a mobile phone, a tablet computer, a personal computer, or a notebook computer.

[0063] It can be learned that, in the solution of this embodiment, after receiving a first voice coded signal from a first terminal supporting a relatively narrow bandwidth, a network device performs voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal. The network device performs, by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal. Then, after combining the first voice decoded signal and the bandwidth extension voice decoded signal, the network device performs voice coding processing to obtain a second voice coded signal, and then sends the second voice coded signal to a second terminal supporting a relatively wide bandwidth. The network device at a transit location performs virtual bandwidth extension on a voice coded signal that is sent by the first terminal supporting a relatively narrow bandwidth to the second terminal supporting a relatively wide bandwidth. Therefore, further, a downlink voice coded signal of the second terminal supporting a relatively wide bandwidth can better match a maximum frequency bandwidth support capability of the second terminal, so that the second terminal supporting a relatively wide bandwidth can enjoy a voice signal bandwidth service that matches the maximum frequency bandwidth support capability of the second terminal. In addition, special function enhancement does not need to be performed on the second terminal, thereby improving call experience of a user.

[0064] The following first provides more detailed description with reference to related accompanying drawings.

[0065] Referring to FIG. 3-A, FIG. 3-A is a schematic flowchart of a voice signal processing method according to an embodiment of this application. As shown in an example in FIG. 3-A, the voice signal processing method provided in this embodiment of this application may include the following steps.

[0066] 301. A call connection is established between a narrowband terminal and a wideband terminal.

[0067] 302. The narrowband terminal performs coding on a voice to obtain a first voice coded signal, and sends the first voice coded signal based on the call connection.

[0068] In this embodiment, it is assumed that a first terminal is the narrowband terminal and a second terminal is the wideband terminal. A maximum frequency bandwidth supported by the narrowband terminal is narrowband (for example, 3400 Hz-300 Hz=3100 Hz), and a maximum frequency bandwidth supported by the wideband terminal is wideband (for example, 7000 Hz-50 Hz=6950 Hz).

[0069] Specifically, for example, the narrowband terminal may perform, by using an AMR NB encoder or another NB encoder, voice coding on a voice sampling signal to obtain a first voice coded signal. A sampling rate of the first voice coded signal is 8 kHz. A frequency bandwidth of the first voice coded signal is BW1=3400 Hz-300 Hz=3100 Hz.

[0070] 303. An RNC receives the first voice coded signal from the narrowband terminal, and the RNC performs voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal.

[0071] For example, the voice decoding parameter may include a pitch period, a voicing factor, and a linear predictive coding parameter.

[0072] It is assumed that a voice decoder group including multiple voice decoders (for example, the voice decoder group may include an NB decoder, a WB decoder, an SWB decoder, and an FB decoder) exists in the RNC. The RNC may select, based on the maximum frequency bandwidth (for example, NB) supported by the narrowband terminal, the NB decoder from the voice decoder group, and perform voice decoding processing on the first voice coded signal to obtain the voice decoding parameter and the first voice decoded signal.

[0073] Referring to FIG. 3-B, for example, the network device (for example, an RNC) in FIG. 3-B includes a voice decoder group (including multiple voice decoders), a VBWE processor group (including multiple VBWE processors), a voice encoder group (including multiple voice encoders), and a down-sampler group (including multiple down-samplers). The RNC may perform a corresponding operation by selecting a corresponding device from a corresponding device group according to a requirement.

[0074] In the embodiments of this application, for example, the NB decoder is an AMR (Adaptive Multi-Rate, adaptive multi-rate)-NB decoder or another type of NB decoder. The SWB decoder may be, for example, an EVS (Enhanced Voice Services, enhanced voice services)-SWB decoder or another type of SWB decoder. The WB decoder is, for example, an AMR-WB decoder or another type of WB decoder. The FB decoder may be, for example, an EVS-FB decoder or another type of FB decoder.

[0075] 304. The RNC performs, by using the voice decoding parameter, VBWE processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal, where the voice decoding parameter may include a pitch period, a voicing factor, a linear predictive coding parameter, and the like.

[0076] It is assumed that a VBWE processor group including multiple VBWE processors (for example, the VBWE processor group may include an NB-VBWE processor, a WB-VBWE processor, an SWB VBWE processor, and an FB VBWE processor) exists in the RNC. The RNC may select, based on the maximum frequency bandwidth (for example, WB) supported by the wideband terminal, the WB VBWE processor from multiple VBWE processors, and perform, by using the voice decoding parameter, VBWE processing to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal.

[0077] For example, that the virtual bandwidth extension processing is performed by using the voice decoding parameter to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal may include: estimating, by using the voice decoding parameter, a bandwidth extension excitation signal corresponding to the first voice decoded signal; estimating, by using the voice decoding parameter, a bandwidth extension spectral envelope corresponding to the first voice decoded signal; and performing, by using a filter corresponding to the bandwidth extension spectral envelope, synthesis processing on the bandwidth extension excitation signal to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal.

[0078] For example, that the bandwidth extension excitation signal corresponding to the first voice decoded signal is estimated by using the voice decoding parameter may include: estimating, by using the voice decoding parameter (for example, a voice decoding parameter such as a pitch period or a voicing factor) and based on a spectrum folding algorithm, a white noise excitation algorithm, or a harmonic noise model algorithm, the bandwidth extension excitation signal corresponding to the first voice decoded signal.

[0079] For example, that the bandwidth extension spectral envelope corresponding to the first voice decoded signal is estimated by using the voice decoding parameter may include: estimating, by using the voice decoding parameter (for example, a voice decoding parameter such as a linear predictive coding parameter) and based on a linear mapping method, a codebook mapping method, or a statistics mapping method, the bandwidth extension spectral envelope corresponding to the first voice decoded signal.

[0080] 305. After combining the first voice decoded signal and the bandwidth extension voice decoded signal, the RNC performs voice coding processing to obtain a second voice coded signal.

[0081] The frequency bandwidth of the first voice coded signal is less than a frequency bandwidth of the second voice coded signal. The sampling rate of the first voice coded signal is less than a sampling rate of the second voice coded signal. For example, the sampling rate of the second voice coded signal is 16 kHz, and the frequency bandwidth of the second voice coded signal is BW2.

[0082] The bandwidth extension voice decoded signal corresponding to the first voice decoded signal includes a high bandwidth extension voice decoded signal whose frequency bandwidth is BWE1=7000 Hz-3400 Hz=3600 Hz. If the bandwidth extension voice decoded signal corresponding to the first voice decoded signal includes the high bandwidth extension voice decoded signal but does not include a low bandwidth extension voice decoded signal, as shown in examples in FIG. 3-C and FIG. 3-D,



[0083] In addition, if the bandwidth extension voice decoded signal corresponding to the first voice decoded signal includes the high bandwidth extension voice decoded signal and the low bandwidth extension voice decoded signal,
BW2=BW1+BWE1+BWE2=3100 Hz+(300 Hz-50 Hz)+(7000 Hz-3400 Hz)=6950 Hz. A frequency bandwidth of a low bandwidth extension voice decoded signal corresponding to the first voice decoded signal is BWE2=(300 Hz-50 Hz)=250 Hz.

[0084] It is assumed that a voice encoder group including multiple voice encoders (for example, the voice encoder group may include an NB encoder, a WB encoder, an SWB encoder, and an FB encoder) exists in the RNC. After combining the first voice decoded signal and the bandwidth extension voice decoded signal, the RNC may select, based on the maximum frequency bandwidth (WB) supported by the wideband terminal, the WB encoder from multiple voice encoders, and perform voice coding processing to obtain the second voice coded signal.

[0085] In the embodiments of this application, the NB encoder may be, for example, an AMR-NB encoder or another type of NB encoder. The SWB encoder may be, for example, an EVS-SWB encoder or another type of SWB encoder. The WB encoder is, for example, an AMR-WB encoder or another type of WB encoder. The FB encoder may be, for example, an EVS-FB encoder or another type of FB encoder.

[0086] 306. The RNC sends the second voice coded signal to the wideband terminal that establishes the call connection to the narrowband terminal.

[0087] Correspondingly, the wideband terminal may receive the second voice coded signal, and perform decoding on the second voice coded signal and play a signal obtained after the decoding.

[0088] 307. The wideband terminal performs coding on a voice to obtain a third voice coded signal, and sends the third voice coded signal based on the call connection.

[0089] For example, a sampling rate of the third voice coded signal is 16 kHz, and a frequency bandwidth of the third voice coded signal is BW3=(7000 Hz-300 Hz)=6700 Hz.

[0090] 308. An RNC receives the third voice coded signal from the wideband terminal, and the RNC performs voice decoding processing on the third voice coded signal to obtain a third voice decoded signal.

[0091] It is assumed that a voice decoder group including multiple voice decoders (for example, the voice decoder group may include an NB decoder, a WB decoder, an SWB decoder, and an FB decoder) exists in the RNC. The RNC may select, based on the maximum frequency bandwidth (WB) supported by the wideband terminal, the WB decoder from multiple voice decoders, and perform voice decoding processing on the third voice coded signal to obtain the third voice decoded signal.

[0092] 309. The RNC performs down-sampling processing on the third voice decoded signal to obtain a fourth voice decoded signal.

[0093] It is assumed that a down-sampler group including multiple down-samplers (for example, the down-sampler group may include an NB down-sampler, a WB down-sampler, an SWB down-sampler, and an FB down-sampler) exists in the RNC. The RNC may select, based on the maximum frequency bandwidth (NB) supported by the narrowband terminal, the NB down-sampler from multiple down-samplers included in the down-sampler group, and perform down-sampling processing on the third voice decoded signal to obtain the fourth voice decoded signal.

[0094] 310. The RNC performs voice coding processing on the fourth voice decoded signal to obtain a fourth voice coded signal.

[0095] It is assumed that a voice encoder group including multiple voice encoders (for example, the voice encoder group may include an NB encoder, a WB encoder, an SWB encoder, and an FB encoder) exists in the RNC. The RNC may select, based on the maximum frequency bandwidth (NB) supported by the narrowband terminal, the NB encoder from multiple voice encoders included in the voice encoder group, and perform voice coding processing on the fourth voice decoded signal to obtain the fourth voice coded signal.

[0096] It can be learned that a frequency bandwidth (NB) of the fourth voice coded signal is less than a frequency bandwidth (WB) of the third voice coded signal, and a sampling rate (8 kHz) of the fourth voice coded signal is less than a sampling rate (16 kHz) of the third voice coded signal.

[0097] The frequency bandwidth of the fourth voice coded signal is BW4=(3400 Hz-300 Hz)=3100 Hz.

[0098] 311. The RNC sends the fourth voice coded signal to the narrowband terminal; or after performing voice enhancement processing on the fourth voice coded signal to obtain a fourth voice coded signal obtained after the voice enhancement processing, the RNC sends the fourth voice coded signal obtained after the voice enhancement processing to the narrowband terminal.

[0099] By means of voice enhancement processing, a gain MOS of the fourth voice coded signal may not be less than a gain MOS of a voice signal that is directly sent by the wideband terminal and whose frequency bandwidth is BW1.

[0100] Product forms of a narrowband terminal and a wideband terminal may be user terminals that have a call function, for example, user equipment (UE).

[0101] It may be understood that there is no definite execution order between step 302 to step 306 and step 307 to step 311.

[0102] Referring to FIG. 3-E, an example in FIG. 3-E shows a flow direction relationship of a voice signal between a narrowband terminal, an RNC (an example of a network device), and a wideband terminal. The narrowband terminal, the RNC, and the wideband terminal may have function devices shown in the example in FIG. 3-E.

[0103] It can be learned that, in the example of the solution of this embodiment, after receiving a narrowband voice coded signal from a narrowband terminal, an RNC performs voice decoding processing on the narrowband voice coded signal to obtain a voice decoding parameter and a narrowband voice decoded signal. The RNC performs, by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the narrowband voice decoded signal. Then, after combining the narrowband voice decoded signal and the bandwidth extension voice decoded signal, the RNC performs voice coding processing to obtain a wideband voice coded signal, and then sends the wideband voice coded signal to a wideband terminal. The RNC at a transit location performs virtual bandwidth extension on a voice coded signal sent by the narrowband terminal to the wideband terminal. Therefore, further, a downlink voice coded signal of the wideband terminal can better match a maximum frequency bandwidth support capability of the wideband terminal, so that the wideband terminal can enjoy a voice signal bandwidth service that matches the maximum frequency bandwidth support capability of the wideband terminal. In addition, special function enhancement does not need to be performed on the wideband terminal, thereby improving call experience of a user.

[0104] Referring to FIG. 4-A, FIG. 4-A is a schematic flowchart of a voice signal processing method according to an embodiment of this application. As shown in an example in FIG. 4-A, the voice signal processing method provided in this embodiment of this application may include the following steps.

[0105] 401. A call connection is established between a wideband terminal and a super wideband terminal.

[0106] 402. The wideband terminal performs coding on a voice to obtain a first voice coded signal, and sends the first voice coded signal based on the call connection.

[0107] In this embodiment, it is assumed that a first terminal is the wideband terminal and a second terminal is the super wideband terminal. A maximum frequency bandwidth supported by the wideband terminal is wideband (for example, 7 kHz-50 Hz=6950 Hz), and a maximum frequency bandwidth supported by the super wideband terminal is super wideband (for example, 14 kHz-50 Hz=13950 Hz).

[0108] Specifically, for example, the wideband terminal may perform, by using an AMR WB encoder or another WB encoder, voice coding on a voice sampling signal to obtain a first voice coded signal. A sampling rate of the first voice coded signal is 16 kHz. A frequency bandwidth of the first voice coded signal is BW1=7 kHz-50 Hz=6950 Hz.

[0109] 403. An RNC receives the first voice coded signal from the wideband terminal, and the RNC performs voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal.

[0110] For example, the voice decoding parameter may include a pitch period, a voicing factor, and a linear predictive coding parameter.

[0111] It is assumed that a voice decoder group including multiple voice decoders (for example, the voice decoder group may include an NB decoder, a WB decoder, an SWB decoder, and an FB decoder) exists in the RNC. The RNC may select, based on the maximum frequency bandwidth (for example, WB) supported by the wideband terminal, the WB decoder from the voice decoder group, and perform voice decoding processing on the first voice coded signal to obtain the voice decoding parameter and the first voice decoded signal.

[0112] 404. The RNC performs, by using the voice decoding parameter, VBWE processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal, where the voice decoding parameter may include a pitch period, a voicing factor, a linear predictive coding parameter, and the like.

[0113] It is assumed that a VBWE processor group including multiple VBWE processors (for example, the VBWE processor group may include an NB-VBWE processor, a WB-VBWE processor, an SWB VBWE processor, and an FB VBWE processor) exists in the RNC. The RNC may select, based on the maximum frequency bandwidth (for example, SWB) supported by the super wideband terminal, the SWB VBWE processor from multiple VBWE processors, and perform, by using the voice decoding parameter, VBWE processing to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal.

[0114] For example, that the virtual bandwidth extension processing is performed by using the voice decoding parameter to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal may include: estimating, by using the voice decoding parameter, a bandwidth extension excitation signal corresponding to the first voice decoded signal; estimating, by using the voice decoding parameter, a bandwidth extension spectral envelope corresponding to the first voice decoded signal; and performing, by using a filter corresponding to the bandwidth extension spectral envelope, synthesis processing on the bandwidth extension excitation signal to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal.

[0115] For example, that the bandwidth extension excitation signal corresponding to the first voice decoded signal is estimated by using the voice decoding parameter may include: estimating, by using the voice decoding parameter (for example, a voice decoding parameter such as a pitch period or a voicing factor) and based on a spectrum folding algorithm, a white noise excitation algorithm, or a harmonic noise model algorithm, the bandwidth extension excitation signal corresponding to the first voice decoded signal.

[0116] For example, that the bandwidth extension spectral envelope corresponding to the first voice decoded signal is estimated by using the voice decoding parameter may include: estimating, by using the voice decoding parameter (for example, a voice decoding parameter such as a linear predictive coding parameter) and based on a linear mapping method, a codebook mapping method, or a statistics mapping method, the bandwidth extension spectral envelope corresponding to the first voice decoded signal.

[0117] 405. After combining the first voice decoded signal and the bandwidth extension voice decoded signal, the RNC performs voice coding processing to obtain a second voice coded signal.

[0118] The frequency bandwidth of the first voice coded signal is less than a frequency bandwidth of the second voice coded signal. The sampling rate of the first voice coded signal is less than a sampling rate of the second voice coded signal. For example, the sampling rate of the second voice coded signal is 32 kHz, and the frequency bandwidth of the second voice coded signal is BW2.

[0119] The bandwidth extension voice decoded signal corresponding to the first voice decoded signal includes a high bandwidth extension voice decoded signal whose frequency bandwidth is BWE1=14 kHz-7 kHz=7 kHz. If the bandwidth extension voice decoded signal corresponding to the first voice decoded signal includes the high bandwidth extension voice decoded signal but does not include a low bandwidth extension voice decoded signal,



[0120] It is assumed that a voice encoder group including multiple voice encoders (for example, the voice encoder group may include an NB encoder, a WB encoder, an SWB encoder, and an FB encoder) exists in the RNC. After combining the first voice decoded signal and the bandwidth extension voice decoded signal, the RNC may select, based on the maximum frequency bandwidth (SWB) supported by the super wideband terminal, the SWB encoder from multiple voice encoders, and perform voice coding processing to obtain the second voice coded signal.

[0121] 406. The RNC sends the second voice coded signal to the super wideband terminal that establishes the call connection to the wideband terminal.

[0122] Correspondingly, the super wideband terminal may receive the second voice coded signal, and perform decoding on the second voice coded signal and play a signal obtained after the decoding.

[0123] 407. The super wideband terminal performs coding on a voice to obtain a third voice coded signal, and sends the third voice coded signal based on the call connection.

[0124] For example, a sampling rate of the third voice coded signal is 32 kHz, and a frequency bandwidth of the third voice coded signal is BW3=(14 kHz-50 Hz)=13950 Hz.

[0125] 408. An RNC receives the third voice coded signal from the super wideband terminal, and the RNC performs voice decoding processing on the third voice coded signal to obtain a third voice decoded signal.

[0126] It is assumed that a voice decoder group including multiple voice decoders (for example, the voice decoder group may include an NB decoder, a WB decoder, an SWB decoder, and an FB decoder) exists in the RNC. The RNC may select, based on the maximum frequency bandwidth (SWB) supported by the super wideband terminal, the SWB decoder from multiple voice decoders, and perform voice decoding processing on the third voice coded signal to obtain the third voice decoded signal.

[0127] 409. The RNC performs down-sampling processing on the third voice decoded signal to obtain a fourth voice decoded signal.

[0128] It is assumed that a down-sampler group including multiple down-samplers (for example, the down-sampler group may include an NB down-sampler, a WB down-sampler, an SWB down-sampler, and an FB down-sampler) exists in the RNC. The RNC may select, based on the maximum frequency bandwidth (WB) supported by the wideband terminal, the WB down-sampler from multiple down-samplers included in the down-sampler group, and perform down-sampling processing on the third voice decoded signal to obtain the fourth voice decoded signal.

[0129] 410. The RNC performs voice coding processing on the fourth voice decoded signal to obtain a fourth voice coded signal.

[0130] It is assumed that a voice encoder group including multiple voice encoders (for example, the voice encoder group may include an NB encoder, a WB encoder, an SWB encoder, and an FB encoder) exists in the RNC. The RNC may select, based on the maximum frequency bandwidth (WB) supported by the wideband terminal, the WB encoder from multiple voice encoders included in the voice encoder group, and perform voice coding processing on the fourth voice decoded signal to obtain the fourth voice coded signal.

[0131] It can be learned that a frequency bandwidth (WB) of the fourth voice coded signal is less than a frequency bandwidth (SWB) of the third voice coded signal, and a sampling rate (16 kHz) of the fourth voice coded signal is less than a sampling rate (32 kHz) of the third voice coded signal.

[0132] The frequency bandwidth of the fourth voice coded signal is BW4=(7000 Hz-500 Hz)=6950 Hz.

[0133] 411. The RNC sends the fourth voice coded signal to the wideband terminal; or after performing voice enhancement processing on the fourth voice coded signal to obtain a fourth voice coded signal obtained after the voice enhancement processing, the RNC sends the fourth voice coded signal obtained after the voice enhancement processing to the wideband terminal.

[0134] By means of voice enhancement processing, a gain MOS of the fourth voice coded signal may not be less than a gain MOS of a voice signal that is directly sent by the wideband terminal and whose frequency bandwidth is BW1.

[0135] Product forms of a wideband terminal and a super wideband terminal may be user terminals that have a call function, for example, user equipment (UE).

[0136] It may be understood that there is no definite execution order between step 402 to step 406 and step 407 to step 411.

[0137] Referring to FIG. 4-B, an example in FIG. 4-B shows a flow direction relationship of a voice signal between a wideband terminal, an RNC (an example of a network device), and a super wideband terminal. The wideband terminal, the RNC, and the super wideband terminal may have function devices shown in the example in FIG. 4-B.

[0138] It can be learned that, in the example of the solution of this embodiment, after receiving a wideband voice coded signal from a wideband terminal, a network device (RNC) performs voice decoding processing on the wideband voice coded signal to obtain a voice decoding parameter and a wideband voice decoded signal. The RNC performs, by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the wideband voice decoded signal. Then, after combining the wideband voice decoded signal and the bandwidth extension voice decoded signal, the RNC performs voice coding processing to obtain a super wideband voice coded signal, and then sends the super wideband voice coded signal to a super wideband terminal. The RNC at a transit location performs virtual bandwidth extension on a voice coded signal sent by the wideband terminal to the super wideband terminal. Therefore, further, a downlink voice coded signal of the super wideband terminal can better match a maximum frequency bandwidth support capability of the super wideband terminal, so that the super wideband terminal can enjoy a voice signal bandwidth service that matches the maximum frequency bandwidth support capability of the super wideband terminal. In addition, special function enhancement does not need to be performed on the super wideband terminal, thereby improving call experience of a user.

[0139] In the example of FIG. 3-A, a first terminal is a narrowband terminal and a second terminal is a wideband terminal; in the example of FIG. 4-A, a first terminal is a wideband terminal and a second terminal is a super wideband terminal. A situation in which a first terminal and a second terminal each are another type of terminal may be learned by analogy.

[0140] Referring to FIG. 5, an embodiment of this application provides a network device 500, including:

a communications interface 510, configured to receive a first voice coded signal from a first terminal;

a first voice decoder 520, configured to perform voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal;

a first virtual bandwidth extension processor 530, configured to perform, by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal; and

a first voice encoder 540, configured to: after combining the first voice decoded signal and the bandwidth extension voice decoded signal, perform voice coding processing to obtain a second voice coded signal, where a frequency bandwidth of the first voice coded signal is less than a frequency bandwidth of the second voice coded signal, and a sampling rate of the first voice coded signal is less than a sampling rate of the second voice coded signal; where

the communications interface 510 is further configured to send the second voice coded signal to a second terminal that establishes a call connection to the first terminal, and a maximum frequency bandwidth supported by the first terminal is less than a maximum frequency bandwidth supported by the second terminal.



[0141] In some possible implementation manners of the present invention, the first virtual bandwidth extension processor 530 is specifically configured to: estimate, by using the voice decoding parameter, a bandwidth extension excitation signal corresponding to the first voice decoded signal; estimate, by using the voice decoding parameter, a bandwidth extension spectral envelope corresponding to the first voice decoded signal; and perform, by using a filter corresponding to the bandwidth extension spectral envelope, synthesis processing on the bandwidth extension excitation signal to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal.

[0142] In some possible implementation manners of the present invention, in a process of estimating, by using the voice decoding parameter, the bandwidth extension excitation signal corresponding to the first voice decoded signal, the first virtual bandwidth extension processor 530 is specifically configured to estimate, by using the voice decoding parameter and based on a spectrum folding algorithm, a white noise excitation algorithm, or a harmonic noise model algorithm, the bandwidth extension excitation signal corresponding to the first voice decoded signal.

[0143] In some possible implementation manners of the present invention, in a process of estimating, by using the voice decoding parameter, the bandwidth extension spectral envelope corresponding to the first voice decoded signal, the first virtual bandwidth extension processor 530 is specifically configured to estimate, by using the voice decoding parameter and based on a linear mapping method, a codebook mapping method, or a statistics mapping method, the bandwidth extension spectral envelope corresponding to the first voice decoded signal.

[0144] In some possible implementation manners of the present invention, the voice decoding parameter includes a pitch period, a voicing factor, and a linear predictive coding parameter.

[0145] In some possible implementation manners of the present invention, the network device includes multiple voice decoders, and the first voice decoder is a voice decoder that is in multiple voice decoders and that is corresponding to the maximum frequency bandwidth supported by the first terminal.

[0146] In some possible implementation manners of the present invention, the network device includes multiple virtual bandwidth extension processors, and the first virtual bandwidth extension processor is a virtual bandwidth extension processor that is in multiple virtual bandwidth extension processors and that is corresponding to the maximum frequency bandwidth supported by the second terminal.

[0147] In some possible implementation manners of the present invention, the network device includes multiple voice encoders, and the first voice encoder is a voice encoder that is in multiple voice encoders and that is corresponding to the maximum frequency bandwidth supported by the second terminal.

[0148] In some possible implementation manners of the present invention, for example, frequency bandwidths of the first voice coded signal and the second voice coded signal are two of the following frequency bandwidths: narrowband, wideband, super wideband, or full band.

[0149] In some possible implementation manners of the present invention, the network device 500 further includes a second voice decoder 550, a second voice encoder 570, and a first down-sampler 560.

[0150] The communications interface 510 is further configured to receive a third voice coded signal from the second terminal.

[0151] The second voice decoder 550 is configured to perform voice decoding processing on the third voice coded signal to obtain a third voice decoded signal.

[0152] The first down-sampler 560 is configured to perform down-sampling processing on the third voice decoded signal to obtain a fourth voice decoded signal.

[0153] The second voice encoder 570 is configured to perform voice coding processing on the fourth voice decoded signal to obtain a fourth voice coded signal, where a frequency bandwidth of the fourth voice coded signal is less than a frequency bandwidth of the third voice coded signal, and a sampling rate of the fourth voice coded signal is less than a sampling rate of the third voice coded signal.

[0154] The communications interface 510 is further configured to send the fourth voice coded signal to the first terminal; or the communications interface 510 is further configured to: after performing voice enhancement processing on the fourth voice coded signal to obtain a fourth voice coded signal obtained after the voice enhancement processing, send the fourth voice coded signal obtained after the voice enhancement processing to the first terminal.

[0155] In some possible implementation manners of the present invention, the network device 500 includes multiple voice decoders, and the second voice decoder 550 is a voice decoder corresponding to the maximum frequency bandwidth supported by the second terminal.

[0156] In some possible implementation manners of the present invention, the network device includes multiple down-samplers, and the first down-sampler is a down-sampler that is in multiple down-samplers and that is corresponding to the maximum frequency bandwidth supported by the first terminal.

[0157] In some possible implementation manners of the present invention, the network device 500 includes multiple voice encoders, and the second voice encoder 570 is a voice encoder corresponding to the maximum frequency bandwidth supported by the first terminal.

[0158] In some possible implementation manners of the present invention, the network device is a base station, a radio network controller, or a core network device.

[0159] Referring to FIG. 6, an embodiment of the present invention provides a network device 600, including:
a storage unit 620, a communications interface 610, and a processor 630 coupled to the storage unit 620 and the communications interface 610. The storage unit 620 is configured to store an instruction, the processor 630 is configured to execute the instruction, and the communications interface 610 is configured to communicate with another device under control of the processor 630. When executing the instruction, the processor 630 may perform, according to the instruction, any voice signal processing method in the foregoing embodiment.

[0160] Specifically, the processor 630 is configured to: receive a first voice coded signal from a first terminal by using the communications interface 610; perform voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal; perform, by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal; after combining the first voice decoded signal and the bandwidth extension voice decoded signal, perform voice coding processing to obtain a second voice coded signal, where a frequency bandwidth of the first voice coded signal is less than a frequency bandwidth of the second voice coded signal, and a sampling rate of the first voice coded signal is less than a sampling rate of the second voice coded signal; and send, by using the communications interface 610, the second voice coded signal to a second terminal that establishes a call connection to the first terminal, and a maximum frequency bandwidth supported by the first terminal is less than a maximum frequency bandwidth supported by the second terminal.

[0161] In some possible implementation manners of the present invention, the processor 630 is specifically configured to: estimate, by using the voice decoding parameter, a bandwidth extension excitation signal corresponding to the first voice decoded signal; estimate, by using the voice decoding parameter, a bandwidth extension spectral envelope corresponding to the first voice decoded signal; and perform, by using a filter corresponding to the bandwidth extension spectral envelope, synthesis processing on the bandwidth extension excitation signal to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal.

[0162] In some possible implementation manners of the present invention, in a process of estimating, by using the voice decoding parameter, the bandwidth extension excitation signal corresponding to the first voice decoded signal, the processor 630 is specifically configured to estimate, by using the voice decoding parameter and based on a spectrum folding algorithm, a white noise excitation algorithm, or a harmonic noise model algorithm, the bandwidth extension excitation signal corresponding to the first voice decoded signal.

[0163] In some possible implementation manners of the present invention, in a process of estimating, by using the voice decoding parameter, the bandwidth extension spectral envelope corresponding to the first voice decoded signal, the processor 630 is specifically configured to estimate, by using the voice decoding parameter and based on a linear mapping method, a codebook mapping method, or a statistics mapping method, the bandwidth extension spectral envelope corresponding to the first voice decoded signal.

[0164] In some possible implementation manners of the present invention, the voice decoding parameter includes a pitch period, a voicing factor, and a linear predictive coding parameter.

[0165] In some possible implementation manners of the present invention, for example, frequency bandwidths of the first voice coded signal and the second voice coded signal are two of the following frequency bandwidths: narrowband, wideband, super wideband, or full band.

[0166] In some possible implementation manners of the present invention, the processor 630 is further configured to: receive a third voice coded signal from the second terminal by using the communications interface 610; perform voice decoding processing on the third voice coded signal to obtain a third voice decoded signal; perform down-sampling processing on the third voice decoded signal to obtain a fourth voice decoded signal; perform voice coding processing on the fourth voice decoded signal to obtain a fourth voice coded signal, where a frequency bandwidth of the fourth voice coded signal is less than a frequency bandwidth of the third voice coded signal, and a sampling rate of the fourth voice coded signal is less than a sampling rate of the third voice coded signal; and send the fourth voice coded signal to the first terminal by using the communications interface 610, or after performing voice enhancement processing on the fourth voice coded signal to obtain a fourth voice coded signal obtained after the voice enhancement processing, send, by using the communications interface 610, the fourth voice coded signal obtained after the voice enhancement processing to the first terminal.

[0167] In some possible implementation manners of the present invention, the network device 600 may be, for example, a base station, a radio network controller, a core network device, or a network telephony server.

[0168] It can be learned that, in the foregoing technical solution, after receiving a first voice coded signal from a first terminal supporting a relatively narrow bandwidth, a network device 600 performs voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal. The network device 600 performs, by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal. Then, after combining the first voice decoded signal and the bandwidth extension voice decoded signal, the network device 600 performs voice coding processing to obtain a second voice coded signal, and then sends the second voice coded signal to a second terminal supporting a relatively wide bandwidth. The network device 600 at a transit location performs virtual bandwidth extension on a voice coded signal that is sent by the first terminal supporting a relatively narrow bandwidth to the second terminal supporting a relatively wide bandwidth. Therefore, further, a downlink voice coded signal of the second terminal supporting a relatively wide bandwidth can better match a maximum frequency bandwidth support capability of the second terminal, so that the second terminal supporting a relatively wide bandwidth can enjoy a voice signal bandwidth service that matches the maximum frequency bandwidth support capability of the second terminal. In addition, special function enhancement does not need to be performed on the second terminal, thereby improving call experience of a user. It can be learned that, in the foregoing example, service quality of terminals that have asymmetric maximum frequency bandwidth support capabilities can be improved.

[0169] In the foregoing embodiments, the description of each embodiment has respective focuses. For a part that is not described in detail in an embodiment, reference may be made to related descriptions in other embodiments.

[0170] In the several embodiments provided in this application, it should be understood that the disclosed apparatus may be implemented in other manners. For example, the described apparatus embodiment is merely an example. For example, the unit division is merely logical function division and may be other division in actual implementation. For example, a plurality of units or components may be combined or integrated into another system, or some features may be ignored or not performed. In addition, the displayed or discussed mutual indirect couplings or direct couplings or communication connections may be implemented by using some interfaces. The indirect couplings or communication connections between the apparatuses or units may be implemented in electronic or other forms.

[0171] The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.

[0172] In addition, function units in the embodiments of the present invention may be integrated into one processing unit, or each of the units may exist alone physically, or two or more units are integrated into one unit. The integrated unit may be implemented in a form of hardware, or may be implemented in a form of a software function unit.

[0173] When the integrated unit is implemented in the form of a software function unit and sold or used as an independent product, the integrated unit may be stored in a computer-readable storage medium. Based on such an understanding, the technical solutions of the present invention essentially, or the part contributing to the prior art, or all or a part of the technical solutions may be implemented in the form of a software product. The software product is stored in a storage medium and includes several instructions for instructing a computer device (which may be a personal computer, a server, or a network device) to perform all or a part of the steps of the methods described in the embodiments of the present invention. The foregoing storage medium includes: any medium that can store program code, such as a USB flash drive, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a removable hard disk, a magnetic disk, or an optical disc.


Claims

1. A voice signal processing method, comprising:

receiving (201, 302, 402), by a network device, a first voice coded signal from a first terminal;

performing (202, 303, 403), by the network device, voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal;

performing (203, 304, 404), by the network device by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal; after combining the first voice decoded signal and the bandwidth extension voice decoded signal, performing (204, 305, 405), by the network device, voice coding processing to obtain a second voice coded signal, wherein a frequency bandwidth of the first voice coded signal is less than a frequency bandwidth of the second voice coded signal, and a sampling rate of the first voice coded signal is less than a sampling rate of the second voice coded signal; and

sending (205, 306, 406), by the network device, the second voice coded signal to a second terminal that establishes a call connection to the first terminal, wherein a maximum frequency bandwidth supported by the first terminal is less than a maximum frequency bandwidth supported by the second terminal,

wherein the performing (203, 304, 404), by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal comprises:

estimating, by using the voice decoding parameter, a bandwidth extension excitation signal corresponding to the first voice decoded signal, wherein the estimating the bandwidth extension excitation signal comprises estimating, by using the voice decoding parameter and based on a harmonic noise model algorithm, the bandwidth extension excitation signal corresponding to the first voice decoded signal; estimating, by using the voice decoding parameter, a bandwidth extension spectral envelope corresponding to the first voice decoded signal; performing, by using a filter corresponding to the bandwidth extension spectral envelope, synthesis processing on the bandwidth extension excitation signal to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal; and

selecting, from multiple virtual bandwidth extension processors, a virtual bandwidth extension processor corresponding to the maximum frequency bandwidth supported by the second terminal, and performing, by using the voice decoding parameter, the virtual bandwidth extension processing using the selected virtual bandwidth extension processor to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal.


 
2. The method according to claim 1, wherein the estimating, by using the voice decoding parameter, a bandwidth extension spectral envelope corresponding to the first voice decoded signal comprises: estimating, by using the voice decoding parameter and based on a linear mapping method, a codebook mapping method, or a statistics mapping method, the bandwidth extension spectral envelope corresponding to the first voice decoded signal.
 
3. The method according to any one of claims 1 to 2, wherein the voice decoding parameter comprises a pitch period, a voicing factor, and a linear predictive coding parameter.
 
4. The method according to any one of claims 1 to 3, wherein the performing (202, 303, 403) voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal comprises: selecting, from multiple voice decoders, a voice decoder corresponding to the maximum frequency bandwidth supported by the first terminal, and performing the voice decoding processing on the first voice coded signal to obtain the voice decoding parameter and the first voice decoded signal; or
after combining the first voice decoded signal and the bandwidth extension voice decoded signal, the performing voice coding processing to obtain a second voice coded signal comprises: selecting, from multiple voice encoders, a voice encoder corresponding to the maximum frequency bandwidth supported by the second terminal, and after combining the first voice decoded signal and the bandwidth extension voice decoded signal, performing the voice coding processing to obtain the second voice coded signal.
 
5. The method according to any one of claims 1 to 4, wherein:
frequency bandwidths of the first voice coded signal and the second voice coded signal are two of the following frequency bandwidths: narrowband, wideband, super wideband, or full band.
 
6. The method according to any one of claims 1 to 5, wherein the voice signal processing method further comprises:

receiving (307, 407), by the network device, a third voice coded signal from the second terminal;

performing (308, 408), by the network device, voice decoding processing on the third voice coded signal to obtain a third voice decoded signal;

performing (309, 409), by the network device, down-sampling processing on the third voice decoded signal to obtain a fourth voice decoded signal; and performing (310, 410), by the network device, voice coding processing on the fourth voice decoded signal to obtain a fourth voice coded signal, wherein a frequency bandwidth of the fourth voice coded signal is less than a frequency bandwidth of the third voice coded signal, and a sampling rate of the fourth voice coded signal is less than a sampling rate of the third voice coded signal; and

sending (311, 411), by the network device, the fourth voice coded signal to the first terminal; or after the network device performs voice enhancement processing on the fourth voice coded signal to obtain a fourth voice coded signal obtained after the voice enhancement processing, sending the fourth voice coded signal obtained after the voice enhancement processing to the first terminal.


 
7. The method according to claim 6, wherein the performing (308, 408) voice decoding processing on the third voice coded signal to obtain a third voice decoded signal comprises: selecting, from multiple voice decoders, a voice decoder corresponding to the maximum frequency bandwidth supported by the second terminal, and
performing the voice decoding processing on the third voice coded signal to obtain the third voice decoded signal; or
the performing down-sampling processing on the third voice decoded signal to obtain a fourth voice decoded signal comprises: selecting, multiple down-samplers, a down-sampler corresponding to the maximum frequency bandwidth supported by the first terminal, and performing the down-sampling processing on the third voice decoded signal to obtain the fourth voice decoded signal; or
the performing voice coding processing on the fourth voice decoded signal to obtain a fourth voice coded signal comprises: selecting, from multiple voice encoders, a voice encoder corresponding to the maximum frequency bandwidth supported by the first terminal, and performing the voice coding processing on the fourth voice decoded signal to obtain the fourth voice coded signal.
 
8. The method according to any one of claims 1 to 7, wherein the network device is a base station, a radio network controller, or a core network device.
 
9. A network device (500), comprising:

a communications interface (510), configured to receive a first voice coded signal from a first terminal;

a first voice decoder (520), configured to perform voice decoding processing on the first voice coded signal to obtain a voice decoding parameter and a first voice decoded signal;

a first virtual bandwidth extension processor (530), configured to perform, by using the voice decoding parameter, virtual bandwidth extension processing to obtain a bandwidth extension voice decoded signal corresponding to the first voice decoded signal; and

a first voice encoder (540), configured to: after combining the first voice decoded signal and the bandwidth extension voice decoded signal, perform voice coding processing to obtain a second voice coded signal, wherein a frequency bandwidth of the first voice coded signal is less than a frequency bandwidth of the second voice coded signal, and a sampling rate of the first voice coded signal is less than a sampling rate of the second voice coded signal; wherein

the communications interface (510) is further configured to send the second voice coded signal to a second terminal that establishes a call connection to the first terminal, and a maximum frequency bandwidth supported by the first terminal is less than a maximum frequency bandwidth supported by the second terminal,

wherein the first virtual bandwidth extension processor (530) is specifically configured to: estimate, by using the voice decoding parameter, a bandwidth extension excitation signal corresponding to the first voice decoded signal,

wherein in a process of estimating, by using the voice decoding parameter, the bandwidth extension excitation signal corresponding to the first voice decoded signal, the first virtual bandwidth extension processor (530) is specifically configured to estimate, by using the voice decoding parameter and based on a harmonic noise model algorithm, the bandwidth extension excitation signal corresponding to the first voice decoded signal;

estimate, by using the voice decoding parameter, a bandwidth extension spectral envelope corresponding to the first voice decoded signal; and perform, by using a filter corresponding to the bandwidth extension spectral envelope, synthesis processing on the bandwidth extension excitation signal to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal,

wherein the network device is further configured to select, from multiple virtual bandwidth extension processors, the first virtual bandwidth extension processor as the virtual bandwidth extension processor corresponding to the maximum frequency bandwidth supported by the second terminal, and perform, by using the voice decoding parameter, the virtual bandwidth extension processing to obtain the bandwidth extension voice decoded signal corresponding to the first voice decoded signal.


 
10. The network device (500) according to claim 9, wherein:
in a process of estimating, by using the voice decoding parameter, the bandwidth extension spectral envelope corresponding to the first voice decoded signal, the first virtual bandwidth extension processor (530) is specifically configured to estimate, by using the voice decoding parameter and based on a linear mapping method, a codebook mapping method, or a statistics mapping method, the bandwidth extension spectral envelope corresponding to the first voice decoded signal.
 
11. The network device (500) according to any one of claims 9 to 10, wherein the voice decoding parameter comprises a pitch period, a voicing factor, and a linear predictive coding parameter.
 
12. The network device (500) according to any one of claims 9 to 11, wherein the network device comprises multiple voice decoders, and the first voice decoder (520) is a voice decoder that is in multiple voice decoders and that is corresponding to the maximum frequency bandwidth supported by the first terminal; or,
the network device comprises multiple virtual bandwidth extension processors, and the first virtual bandwidth extension processor (530) is a virtual bandwidth extension processor that is in multiple virtual bandwidth extension processors and that is corresponding to the maximum frequency bandwidth supported by the second terminal; or
the network device comprises multiple voice encoders, and the first voice encoder (540) is a voice encoder that is in multiple voice encoders and that is corresponding to the maximum frequency bandwidth supported by the second terminal.
 
13. The network device (500) according to any one of claims 9 to 12, wherein frequency bandwidths of the first voice coded signal and the second voice coded signal are two of the following frequency bandwidths: narrowband, wideband, super wideband, or full band.
 
14. The network device (500) according to any one of claims 9 to 13, wherein the network device further comprises a second voice decoder (550), a second voice encoder (570), and a first down-sampler (560);
the communications interface (510) is further configured to receive a third voice coded signal from the second terminal;
the second voice decoder (550) is configured to perform voice decoding processing on the third voice coded signal to obtain a third voice decoded signal;
the first down-sampler (560) is configured to perform down-sampling processing on the third voice decoded signal to obtain a fourth voice decoded signal;
the second voice encoder (570) is configured to perform voice coding processing on the fourth voice decoded signal to obtain a fourth voice coded signal, wherein a frequency bandwidth of the fourth voice coded signal is less than a frequency bandwidth of the third voice coded signal, and a sampling rate of the fourth voice coded signal is less than a sampling rate of the third voice coded signal; and
the communications interface (510) is further configured to send the fourth voice coded signal to the first terminal; or the communications interface is further configured to: after performing voice enhancement processing on the fourth voice coded signal to obtain a fourth voice coded signal obtained after the voice enhancement processing, send the fourth voice coded signal obtained after the voice enhancement processing to the first terminal.
 
15. The network device (500) according to claim 14, wherein the network device comprises multiple voice decoders, and the second voice decoder (550) is a voice decoder corresponding to the maximum frequency bandwidth supported by the second terminal; or
the network device comprises multiple down-samplers, and the first down-sampler (560) is a down-sampler that is in multiple down-samplers and that is corresponding to the maximum frequency bandwidth supported by the first terminal; or
the network device comprises multiple voice encoders, and the second voice encoder (570) is a voice encoder corresponding to the maximum frequency bandwidth supported by the first terminal.
 
16. The network device (500) according to any one of claims 9 to 15, wherein the network device is a base station, a radio network controller, or a core network device.
 
17. A communications system, comprising the network device (500) according to any one of claims 9 to 16.
 


Ansprüche

1. Sprachsignalverarbeitungsverfahren, umfassend:

Empfangen (201, 302, 402) eines ersten sprachcodierten Signals von einem ersten Endgerät durch eine Netzwerkvorrichtung;

Durchführen (202, 303, 403) einer Sprachdekodierungsverarbeitung auf dem ersten sprachcodierten Signal durch die Netzwerkvorrichtung, um einen Sprachdekodierungsparameter und ein erstes sprachdekodiertes Signal zu erhalten;

Durchführen (203, 304, 404) einer Verarbeitung der virtuellen Bandbreitenerweiterung durch die Netzwerkvorrichtung unter Verwendung des Sprachdekodierungsparameters, um ein dem ersten sprachdekodierten Signal entsprechendes sprachdekodiertes Signal der Bandbreitenerweiterung zu erhalten; nach dem Kombinieren des ersten sprachdekodierten Signals und des sprachdekodierten Bandbreitenerweiterungssignals, Durchführen (204, 305, 405) einer Sprachcodierungsverarbeitung durch die Netzwerkkarte, um ein zweites sprachkodiertes Signal zu erhalten, wobei eine Frequenzbandbreite des ersten sprachkodierten Signals kleiner als eine Frequenzbandbreite des zweiten sprachkodierten Signals ist, und eine Abtastrate des ersten sprachkodierten Signals kleiner als eine Abtastrate des zweiten sprachkodierten Signals ist; und

Senden (205, 306, 406) des zweiten sprachcodierten Signals durch die Netzwerkvorrichtung an ein zweites Endgerät, das eine Anrufverbindung zu dem ersten Endgerät herstellt, wobei eine von dem ersten Endgerät unterstützte maximale Frequenzbandbreite kleiner als eine von dem zweiten Endgerät unterstützte maximale Frequenzbandbreite ist,

wobei das Durchführen (203, 304, 404) einer virtuellen Bandbreitenerweiterungsverarbeitung unter Verwendung des Sprachdekodierungsparameters zum Erhalten eines sprachdecodierten Bandbreitenerweiterungssignals, das dem ersten Sprachdekodierungssignal entspricht, umfasst: Schätzen eines Bandbreitenerweiterungsanregungssignals, das dem ersten Sprachdekodierungssignal entspricht, unter Verwendung des Sprachdekodierungsparameters, wobei das Schätzen des Bandbreitenerweiterungsanregungssignals das Schätzen des Bandbreitenerweiterungsanregungssignals entsprechend dem ersten sprachdekodierten Signal unter Verwendung des Sprachdekodierungsparameters und basierend auf einem harmonischen Rauschmodellalgorithmus umfasst; das Schätzen, unter Verwendung des Sprachdekodierungsparameters, einer spektralen Hüllkurve zur Bandbreitenerweiterung entsprechend dem ersten sprachdekodierten Signal; das Durchführen einer Syntheseverarbeitung des Bandbreitenerweiterungsanregungssignals unter Verwendung eines Filters entsprechend der spektralen Hüllkurve zur Bandbreitenerweiterung, um das sprachdecodierte Bandbreitenerweiterungssignal entsprechend dem ersten sprachdekodierten Signal zu erhalten; und

Auswählen eines virtuellen Bandbreitenerweiterungsprozessors aus mehreren virtuellen Bandbreitenerweiterungsprozessoren, der der vom zweiten Endgerät unterstützten maximalen Frequenzbandbreite entspricht, und Durchführen der Verarbeitung der virtuellen Bandbreitenerweiterung unter Verwendung des Sprachdekodierungsparameters unter Verwendung des ausgewählten virtuellen Bandbreitenerweiterungsprozessors, um das dem ersten sprachdekodierten Signal entsprechende sprachdekodierte Bandbreitenerweiterungssignal zu erhalten.


 
2. Verfahren nach Anspruch 1, wobei das Schätzen einer spektralen Hüllkurve zur Bandbreitenerweiterung, die dem ersten sprachdekodierten Signal entspricht, unter Verwendung der Sprachdecodierungsparameter umfasst: Schätzen der spektralen Hüllkurve zur Bandbreitenerweiterung, die dem ersten sprachdekodierten Signal entspricht, unter Verwendung des Sprachdekodierungsparameters und basierend auf einem linearen Zuordnungsverfahren, einem Codebuchzuordnungsverfahren oder einem Statistikzuordnungsverfahren.
 
3. Verfahren nach einem der Ansprüche 1 bis 2, wobei der Sprachdekodierungsparameter eine Tonhöhenperiode, einen Sprachfaktor und einen linearen vorhersagefähigen Kodierungsparameter umfasst.
 
4. Verfahren nach einem der Ansprüche 1 bis 3, wobei das Durchführen (202, 303, 403) einer Sprachdekodierungsverarbeitung auf dem ersten sprachcodierten Signal zum Erhalten eines Sprachdekodierungsparameters und eines ersten sprachdekodierten Signals umfasst: Auswählen eines Sprachdekodierers aus mehreren Sprachdekodierern, der der maximalen Frequenzbandbreite entspricht, die von dem ersten Endgerät unterstützt wird, und Durchführen der Sprachdekodierungsverarbeitung auf dem ersten sprachcodierten Signal zum Erhalten des Sprachdekodierungsparameters und des ersten sprachdekodierten Signals; oder
nach dem Kombinieren des ersten sprachdekodierten Signals und des sprachdekodierten Bandbreitenerweiterungssignals, wobei die Durchführung der Sprachcodierungsverarbeitung zum Erhalten eines zweiten sprachkodierten Signals umfasst: Auswählen eines Sprachcodierers, der der vom zweiten Endgerät unterstützten maximalen Frequenzbandbreite entspricht, aus mehreren Sprachcodierern, und nach dem Kombinieren des ersten sprachdekodierten Signals und des sprachdekodierten Bandbreitenerweiterungssignals, Durchführen der Sprachcodierungsverarbeitung zum Erhalten des zweiten sprachkodierten Signals.
 
5. Verfahren nach einem der Ansprüche 1 bis 4, wobei
Frequenzbandbreiten des ersten sprachcodierten Signals und des zweiten sprachcodierten Signals zwei der folgenden Frequenzbandbreiten sind: Schmalband, Breitband, Superbreitband oder Vollband.
 
6. Verfahren nach einem der Ansprüche 1 bis 5, wobei das Sprachsignalverarbeitungsverfahren ferner umfasst:

Empfangen (307, 407) eines dritten sprachcodierten Signals von dem zweiten Endgerät durch die Netzwerkvorrichtung;

Durchführen (308, 408) einer Sprachdekodierungsverarbeitung auf dem dritten sprachcodierten Signal durch die Netzwerkvorrichtung, um ein drittes sprachdekodiertes Signal zu erhalten;

Durchführen (309, 409) einer Downsampling-Verarbeitung auf dem dritten sprachdekodierten Signal durch die Netzwerkvorrichtung, um ein viertes sprachdekodiertes Signal zu erhalten; und Durchführen (310, 410) einer Sprachkodierungsverarbeitung auf dem vierten sprachdekodierten Signal durch die Netzwerkvorrichtung, um ein viertes sprachkodiertes Signal zu erhalten, wobei eine Frequenzbandbreite des vierten sprachkodierten Signals kleiner als eine Frequenzbandbreite des dritten sprachkodierten Signals ist, und eine Abtastrate des vierten sprachkodierten Signals kleiner als eine Abtastrate des dritten sprachkodierten Signals ist; und

Senden (311, 411) des vierten sprachcodierten Signals durch die Netzwerkvorrichtung an das erste Endgerät; oder nachdem die Netzwerkvorrichtung eine Sprachverbesserungsverarbeitung auf dem vierten sprachcodierten Signal durchgeführt hat, um ein viertes sprachcodiertes Signal zu erhalten, das nach der Sprachverbesserungsverarbeitung erhalten wurde, Senden des vierten sprachcodierten Signals, das nach der Sprachverbesserungsverarbeitung erhalten wurde, an das erste Endgerät.


 
7. Verfahren nach Anspruch 6, wobei die Durchführung (308, 408) der Sprachdekodierungsverarbeitung auf dem dritten sprachkodierten Signal zum Erhalten eines dritten sprachdekodierten Signals umfasst: Auswählen eines Sprachdekodierers aus mehreren Sprachdekodierern, der der maximalen Frequenzbandbreite entspricht, die von dem zweiten Endgerät unterstützt wird, und Durchführen der Sprachdekodierungsverarbeitung auf dem dritten sprachkodierten Signal zum Erhalten des dritten sprachdekodierten Signals; oder
die Durchführung einer Downsampling-Verarbeitung auf dem dritten sprachdekodierten Signal, um ein viertes sprachdekodiertes Signal zu erhalten, umfasst: Auswählen mehrerer Downsampler, eines Downsamplers, der der maximalen Frequenzbandbreite entspricht, die von dem ersten Endgerät unterstützt wird, und Ausführen der Downsampling-Verarbeitung auf dem dritten sprachdekodierten Signal, um das vierte sprachdekodierte Signal zu erhalten; oder
das Durchführen einer Sprachcodierungsverarbeitung auf dem vierten sprachdecodierten Signal, um ein viertes sprachcodiertes Signal zu erhalten, umfasst: Auswählen eines Sprachcodierers aus mehreren Sprachcodierern, der der maximalen Frequenzbandbreite entspricht, die von dem ersten Endgerät unterstützt wird, und Durchführen der Sprachcodierungsverarbeitung auf dem vierten sprachdecodierten Signal, um das vierte sprachcodierte Signal zu erhalten.
 
8. Verfahren nach einem der Ansprüche 1 bis 7, wobei die Netzwerkvorrichtung eine Basisstation, eine Funknetzwerksteuerung oder eine Kemnetzwerkvorrichtung ist.
 
9. Netzwerkvorrichtung (500), umfassend:

eine Kommunikationsschnittstelle (510), die zum Empfangen eines ersten sprachcodierten Signals von einem ersten Endgerät ausgelegt ist;

einen ersten Sprachdecoder (520), der zum Durchführen einer Sprachdecodierungsverarbeitung auf dem ersten sprachcodierten Signal ausgelegt ist, um einen Sprachdecodierungsparameter und ein erstes sprachdecodiertes Signal zu erhalten;

einen ersten Prozessor (530) zur Erweiterung der virtuellen Bandbreite, der zum Durchführen einer Verarbeitung zur Erweiterung der virtuellen Bandbreite unter Verwendung des Sprachdekodierungsparameters ausgelegt ist, um ein dem ersten sprachdekodierten Signal entsprechendes sprachdekodiertes Signal zur Erweiterung der Bandbreite zu erhalten; und

einen ersten Sprachcodierer (540), der ausgelegt ist zum: nach dem Kombinieren des ersten sprachdecodierten Signals und des sprachdecodierten Bandbreitenerweiterungssignals, Durchführen einer Sprachcodierung, um ein zweites sprachcodiertes Signal zu erhalten, wobei eine Frequenzbandbreite des ersten sprachcodierten Signals kleiner als eine Frequenzbandbreite des zweiten sprachcodierten Signals ist, und eine Abtastrate des ersten sprachcodierten Signals kleiner als eine Abtastrate des zweiten sprachcodierten Signals ist; wobei

die Kommunikationsschnittstelle (510) ferner zum Senden des zweiten sprachcodierten Signals an ein zweites Endgerät ausgelegt ist, das eine Anrufverbindung zu dem ersten Endgerät herstellt, und eine von dem ersten Endgerät unterstützte maximale Frequenzbandbreite kleiner als eine von dem zweiten Endgerät unterstützte maximale Frequenzbandbreite ist,

wobei der erste virtuelle Bandbreitenerweiterungsprozessor (530) insbesondere ausgelegt ist zum: Schätzen eines Bandbreitenerweiterungsanregungssignals, das dem ersten sprachdekodierten Signal entspricht, unter Verwendung des Sprachdekodierungsparameters, wobei

in einem Prozess des Schätzens des Bandbreitenerweiterungsanregungssignals, das dem ersten sprachdekodierten Signal entspricht, unter Verwendung des Sprachdekodierungsparameters, der erste virtuelle Bandbreitenerweiterungsprozessor (530) insbesondere zum Schätzen des Bandbreitenerweiterungsanregungssignals, das dem ersten sprachdekodierten Signal entspricht, unter Verwendung des Sprachdekodierungsparameters und basierend auf einem harmonischen Rauschmodellalgorithmus, ausgelegt ist;

Schätzen, unter Verwendung des Sprachdekodierungsparameters, einer spektralen Hüllkurve zur Bandbreitenerweiterung, die dem ersten sprachdekodierten Signal entspricht; und unter Verwendung eines Filters, der der spektralen Hüllkurve zur Bandbreitenerweiterung entspricht, eine Syntheseverarbeitung des Bandbreitenerweiterungsanregungssignals durchzuführen, um das dem ersten sprachdekodierten Signal entsprechende sprachdekodierte Signal der Bandbreitenerweiterung zu erhalten, wobei die Netzwerkvorrichtung ferner zum Auswählen des ersten virtuellen Bandbreitenerweiterungsprozessors als virtuellen Bandbreitenerweiterungsprozessors entsprechend der vom zweiten Endgerät unterstützten maximalen Frequenzbandbreite aus mehreren virtuellen Bandbreitenerweiterungsprozessoren ausgelegt ist, und unter Verwendung des Sprachdekodierungsparameters die Verarbeitung der virtuellen Bandbreitenerweiterung durchzuführen, um das dem ersten sprachdekodierten Signal entsprechende sprachdekodierte Signal der Bandbreitenerweiterung zu erhalten.


 
10. Netzwerkvorrichtung (500) nach Anspruch 9, wobei:
in einem Prozess des Schätzens der spektralen Hüllkurve zur Bandbreitenerweiterung, die dem ersten sprachdekodierten Signal entspricht, unter Verwendung des Sprachdekodierungsparameters und basierend auf einem linearen Zuordnungsverfahren, einem Codebuchzuordnungsverfahren oder einem Statistikzuordnungsverfahren, der erste virtuelle Bandbreitenerweiterungsprozess (530) insbesondere zum Schätzen der spektralen Hüllkurve zur Bandbreitenerweiterung entsprechend dem ersten sprachdekodierten Signal ausgelegt ist.
 
11. Netzwerkvorrichtung (500) nach einem der Ansprüche 9 bis 10, wobei der Sprachdekodierungsparameter eine Tonhöhenperiode, einen Sprachfaktor und einen linearen vorhersagefähigen Kodierungsparameter umfasst.
 
12. Netzwerkvorrichtung (500) nach einem der Ansprüche 9 bis 11, wobei die Netzwerkvorrichtung mehrere Sprachdecoder umfasst, und der erste Sprachdecoder (520) ein Sprachdecoder ist, der sich in mehreren Sprachdecodern befindet und der der maximalen Frequenzbandbreite entspricht, die von dem ersten Endgerät unterstützt wird; oder,
die Netzwerkvorrichtung mehrere Prozessoren zur Erweiterung der virtuellen Bandbreite umfasst, und der erste Prozessor (530) zur Erweiterung der virtuellen Bandbreite ein Prozessor zur Erweiterung der virtuellen Bandbreite ist, der sich in mehreren Prozessoren zur Erweiterung der virtuellen Bandbreite befindet und der der maximalen Frequenzbandbreite entspricht, die von dem zweiten Endgerät unterstützt wird; oder
die Netzwerkvorrichtung mehrere Sprachkodierer umfasst, und der erste Sprachkodierer (540) ein Sprachkodierer ist, der sich in mehreren Sprachkodierern befindet und der der maximalen Frequenzbandbreite entspricht, die von dem zweiten Endgerät unterstützt wird.
 
13. Netzwerkvorrichtung (500) nach einem der Ansprüche 9 bis 12, wobei die Frequenzbandbreiten des ersten sprachcodierten Signals und des zweiten sprachcodierten Signals zwei der folgenden Frequenzbandbreiten sind: Schmalband, Breitband, Superbreitband oder Vollband.
 
14. Die Netzwerkvorrichtung (500) nach einem der Ansprüche 9 bis 13, wobei die Netzwerkvorrichtung ferner einen zweiten Sprachdecoder (550), einen zweiten Sprachcodierer (570) und einen ersten Downsampler (560) umfasst;
die Kommunikationsschnittstelle (510) ferner zum Empfangen eines dritten sprachcodierten Signals von dem zweiten Endgerät ausgelegt ist;
der zweite Sprachdecoder (550) zum Durchführen einer Sprachdecodierungsverarbeitung auf dem dritten sprachcodierten Signal ausgelegt ist, um ein drittes sprachdecodiertes Signal zu erhalten;
der erste Downsampler (560) zum Durchführen einer Downsampling-Verarbeitung auf dem dritten sprachdekodierte Signal ausgelegt ist, um ein viertes sprachdekodiertes Signal zu erhalten;
der zweite Sprachcodierer (570) zum Durchführen einer Sprachcodierungsverarbeitung auf dem vierten sprachdecodierten Signal ausgelegt ist, um ein viertes sprachcodiertes Signal zu erhalten, wobei eine Frequenzbandbreite des vierten sprachcodierten Signals kleiner als eine Frequenzbandbreite des dritten sprachcodierten Signals ist, und eine Abtastrate des vierten sprachcodierten Signals kleiner als eine Abtastrate des dritten sprachcodierten Signals ist; und
die Kommunikationsschnittstelle (510) ferner zum Senden des vierten sprachcodierten Signals an das erste Endgerät ausgelegt ist; oder die Kommunikationsschnittstelle ferner zum Senden des vierten sprachcodierten Signals an das erste Endgerät ausgelegt ist: nach dem Durchführen einer Sprachverbesserungsverarbeitung auf dem vierten sprachcodierten Signal zum Erhalten eines vierten sprachcodierten Signals, das nach der Sprachverbesserungsverarbeitung erhalten wurde, Senden des vierten sprachcodierten Signals, das nach der Sprachverbesserungsverarbeitung erhalten wurde, an das erste Endgerät.
 
15. Netzwerkvorrichtung (500) nach Anspruch 14, wobei die Netzwerkvorrichtung mehrere Sprachdecoder umfasst und der zweite Sprachdecoder (550) ein Sprachdecoder ist, der der maximalen Frequenzbandbreite entspricht, die von dem zweiten Endgerät unterstützt wird; oder
die Netzwerkvorrichtung mehrere Downsampler umfasst, und der erste Downsampler (560) ein Downsampler ist, der sich in mehreren Downsamplern befindet und der der maximalen Frequenzbandbreite entspricht, die von dem ersten Endgerät unterstützt wird; oder
die Netzwerkvorrichtung mehrere Sprachcodierer umfasst, und der zweite Sprachcodierer (570) ein Sprachcodierer ist, der der maximalen Frequenzbandbreite entspricht, die von dem ersten Endgerät unterstützt wird.
 
16. Netzwerkvorrichtung (500) nach einem der Ansprüche 9 bis 15, wobei die Netzwerkvorrichtung eine Basisstation, eine Funknetzwerksteuerung oder eine Kemnetzwerkeinrichtung ist.
 
17. Kommunikationssystem, umfassend die Netzwerkvorrichtung (500) nach einem der Ansprüche 9 bis 16.
 


Revendications

1. Procédé de traitement de signal vocal consistant :

à recevoir (201, 302, 402), au moyen d'un dispositif de réseau, un premier signal codé vocalement en provenance d'un premier terminal ;

à effectuer (202, 303, 403), au moyen du dispositif de réseau, un traitement de décodage vocal sur le premier signal codé vocalement pour obtenir un paramètre de décodage vocal et un premier signal décodé vocalement ;

à effectuer (203, 304, 404), au moyen du dispositif de réseau en utilisant le paramètre de décodage vocal, un traitement d'extension de largeur de bande virtuelle pour obtenir un signal décodé vocalement d'extension de largeur de bande correspondant au premier signal décodé vocalement ; après la combinaison du premier signal décodé vocalement et du signal décodé vocalement d'extension de largeur de bande,

à effectuer (204, 305, 405), au moyen du dispositif de réseau, un traitement de codage vocal pour obtenir un deuxième signal codé vocalement, dans lequel une largeur de bande de fréquence du premier signal codé vocalement est inférieure à une largeur de bande de fréquence du deuxième signal codé vocalement et un taux d'échantillonnage du premier signal codé vocalement est inférieur à un taux d'échantillonnage du deuxième signal codé vocalement ; et

à envoyer (205, 306, 406), au moyen du dispositif de réseau, le deuxième signal codé vocalement à un second terminal qui établit une connexion d'appel avec le premier terminal, dans lequel une largeur de bande de fréquence maximale prise en charge par le premier terminal est inférieure à une largeur de bande de fréquence maximale prise en charge par le second terminal,

dans lequel la réalisation (203, 304, 404), en utilisant le paramètre de décodage vocal, d'un traitement d'extension de largeur de bande virtuelle pour obtenir un signal décodé vocalement d'extension de largeur de bande correspondant au premier signal décodé vocalement consiste :

à estimer, en utilisant le paramètre de décodage vocal, un signal d'excitation d'extension de largeur de bande correspondant au premier signal décodé vocalement, dans lequel l'estimation du signal d'excitation d'extension de largeur de bande consiste à estimer, en utilisant le paramètre de décodage vocal et en se basant sur un algorithme de modèle de bruit harmonique, le signal d'excitation d'extension de largeur de bande correspondant au premier signal décodé vocalement ; à estimer, en utilisant le paramètre de décodage vocal, une enveloppe spectrale d'extension de largeur de bande correspondant au premier signal décodé vocalement ; à effectuer, en utilisant un filtre correspondant à l'enveloppe spectrale d'extension de largeur de bande, un traitement de synthèse sur le signal d'excitation d'extension de largeur de bande pour obtenir le signal décodé vocalement d'extension de largeur de bande correspondant au premier signal décodé vocalement ; et

à sélectionner, parmi de multiples processeurs d'extension de largeur de bande virtuelle, un processeur d'extension de largeur de bande virtuelle correspondant à la largeur de bande de fréquence maximale prise en charge par le second terminal et à effectuer, en utilisant le paramètre de décodage vocal, le traitement d'extension de largeur de bande virtuelle à l'aide du processeur d'extension de largeur de bande virtuelle sélectionné pour obtenir le signal décodé vocalement d'extension de largeur de bande correspondant au premier signal décodé vocalement.


 
2. Procédé selon la revendication 1, dans lequel l'estimation, en utilisant le paramètre de décodage vocal, d'une enveloppe spectrale d'extension de largeur de bande correspondant au premier signal décodé vocalement consiste : à estimer, en utilisant le paramètre de décodage vocal et en se basant sur un procédé de mappage linéaire, sur un procédé de mappage de livre de codes ou sur un procédé de mappage de statistiques, l'enveloppe spectrale d'extension de largeur de bande correspondant au premier signal décodé vocalement.
 
3. Procédé selon l'une quelconque des revendications 1 et 2, dans lequel le paramètre de décodage vocal comprend une période de hauteur tonale, un facteur de voisement et un paramètre de codage prédictif linéaire.
 
4. Procédé selon l'une quelconque des revendications 1 à 3, dans lequel la réalisation (202, 303, 403) d'un traitement de décodage vocal sur le premier signal codé vocalement pour obtenir un paramètre de décodage vocal et un premier signal décodé vocalement consiste : à sélectionner, parmi de multiples décodeurs vocaux, un décodeur vocal correspondant à la largeur de bande de fréquence maximale prise en charge par le premier terminal, et à effecteur le traitement de décodage vocal sur le premier signal codé vocalement pour obtenir le paramètre de décodage vocal et le premier signal décodé vocalement ; ou
après la combinaison du premier signal décodé vocalement et du signal décodé vocalement d'extension de largeur de bande, la réalisation du traitement de codage vocal pour obtenir un deuxième signal codé vocalement consiste : à sélectionner, parmi de multiples codeurs vocaux, un codeur vocal correspondant à la largeur de bande de fréquence maximale prise en charge par le second terminal, et, après la combinaison du premier signal décodé vocalement et du signal décodé vocalement d'extension de largeur de bande, à effectuer le traitement de codage vocal pour obtenir le deuxième signal codé vocalement.
 
5. Procédé selon l'une quelconque des revendications 1 à 4, dans lequel :
des largeurs de bande de fréquence du premier signal codé vocalement et du deuxième signal codé vocalement sont deux des largeurs de bande de fréquence suivantes : à bande étroite, à large bande, à super large bande ou à bande complète.
 
6. Procédé selon l'une quelconque des revendications 1 à 5, dans lequel le procédé de traitement de signal vocal consiste en outre :

à recevoir (307, 407), au moyen du dispositif de réseau, un troisième signal codé vocalement en provenance du second terminal ;

à effectuer (308, 408), au moyen du dispositif de réseau, un traitement de décodage vocal sur le troisième signal codé vocalement pour obtenir un troisième signal décodé vocalement ;

à effectuer (309, 409), au moyen du dispositif de réseau, un traitement de sous-échantillonnage sur le troisième signal décodé vocalement pour obtenir un quatrième signal décodé vocalement ; et à effectuer (310, 410), au moyen du dispositif de réseau, un traitement de codage vocal sur le quatrième signal décodé vocalement pour obtenir un quatrième signal codé vocalement, dans lequel une largeur de bande de fréquence du quatrième signal codé vocalement est inférieure à une largeur de bande de fréquence du troisième signal codé vocalement et un taux d'échantillonnage du quatrième signal codé vocalement est inférieur à un taux d'échantillonnage du troisième signal codé vocalement ; et

à envoyer (311, 411), au moyen du dispositif de réseau, le quatrième signal codé vocalement au premier terminal ; ou après que le dispositif de réseau effectue un traitement d'amélioration de la voix sur le quatrième signal codé vocalement pour obtenir un quatrième signal codé vocalement obtenu après le traitement d'amélioration de la voix, à envoyer le quatrième signal codé vocalement obtenu après le traitement d'amélioration de la voix au premier terminal.


 
7. Procédé selon la revendication 6, dans lequel la réalisation (308, 408) d'un traitement de décodage vocal sur le troisième signal codé vocalement pour obtenir un troisième signal décodé vocalement consiste : à sélectionner, parmi de multiples décodeurs vocaux, un décodeur vocal correspondant à la largeur de bande de fréquence maximale prise en charge par le second terminal, et à effecteur le traitement de décodage vocal sur le troisième signal codé vocalement pour obtenir le troisième signal décodé vocalement ; ou
la réalisation d'un traitement de sous-échantillonnage sur le troisième signal décodé vocalement pour obtenir un quatrième signal décodé vocalement consiste : à sélectionner, de multiples dispositifs de sous-échantillonnage, un dispositif de sous-échantillonnage correspondant à la largeur de bande de fréquence maximale prise en charge par le premier terminal, et à effectuer le traitement de sous-échantillonnage sur le troisième signal décodé vocalement pour obtenir le quatrième signal décodé vocalement ; ou
la réalisation d'un traitement de codage vocal sur le quatrième signal décodé vocalement pour obtenir un quatrième signal codé vocalement consiste : à sélectionner, parmi de multiples codeurs vocaux, un codeur vocal correspondant à la largeur de bande de fréquence maximale prise en charge par le premier terminal, et à effectuer le traitement de codage vocal sur le quatrième signal décodé vocalement pour obtenir le quatrième signal codé vocalement.
 
8. Procédé selon l'une quelconque des revendications 1 à 7, dans lequel le dispositif de réseau est une station de base, un dispositif de commande de réseau radio ou un dispositif de réseau central.
 
9. Dispositif de réseau (500) comprenant :

une interface de communication (510), configurée pour recevoir un premier signal codé vocalement en provenance d'un premier terminal ;

un premier décodeur vocal (520), configuré pour effectuer un traitement de décodage vocal sur le premier signal codé vocalement pour obtenir un paramètre de décodage vocal et un premier signal décodé vocalement ;

un premier processeur d'extension de largeur de bande virtuelle (530), configuré pour effectuer, en utilisant le paramètre de décodage vocal, un traitement d'extension de largeur de bande virtuelle pour obtenir un signal décodé vocalement d'extension de largeur de bande correspondant au premier signal décodé vocalement ; et

un premier codeur vocal (540), configuré : après la combinaison du premier signal décodé vocalement et du signal décodé vocalement d'extension de largeur de bande, pour effectuer un traitement de codage vocal pour obtenir un deuxième signal codé vocalement, dans lequel une largeur de bande de fréquence du premier signal codé vocalement est inférieure à une largeur de bande de fréquence du deuxième signal codé vocalement et un taux d'échantillonnage du premier signal codé vocalement est inférieur à un taux d'échantillonnage du deuxième signal codé vocalement ; dans lequel

l'interface de communication (510) est en outre configurée pour envoyer le deuxième signal codé vocalement à un second terminal qui établit une connexion d'appel avec le premier terminal, et une largeur de bande de fréquence maximale prise en charge par le premier terminal est inférieure à une largeur de bande de fréquence maximale prise en charge par le second terminal,

dans lequel le premier processeur d'extension de largeur de bande virtuelle (530) est spécialement configuré : pour estimer, en utilisant le paramètre de décodage vocal, un signal d'excitation d'extension de largeur de bande correspondant au premier signal décodé vocalement,

dans lequel, dans un processus d'estimation, en utilisant le paramètre de décodage vocal, du signal d'excitation d'extension de largeur de bande correspondant au premier signal décodé vocalement, le premier processeur d'extension de largeur de bande virtuelle (530) est spécialement configuré pour estimer, en utilisant le paramètre de décodage vocal et en se basant sur un algorithme de modèle de bruit harmonique, le signal d'excitation d'extension de largeur de bande correspondant au premier signal décodé vocalement ;

pour estimer, en utilisant le paramètre de décodage vocal, une enveloppe spectrale d'extension de largeur de bande correspondant au premier signal décodé vocalement ; et pour effectuer, en utilisant un filtre correspondant à l'enveloppe spectrale d'extension de largeur de bande, un traitement de synthèse sur le signal d'excitation d'extension de largeur de bande pour obtenir le signal décodé vocalement d'extension de largeur de bande correspondant au premier signal décodé vocalement ;

dans lequel le dispositif de réseau est en outre configuré pour sélectionner, parmi de multiples processeurs d'extension de largeur de bande virtuelle le premier processeur d'extension de largeur de bande virtuelle en tant que processeur d'extension de largeur de bande virtuelle correspondant à la largeur de bande de fréquence maximale prise en charge par le second terminal et pour effectuer, en utilisant le paramètre de décodage vocal, le traitement d'extension de largeur de bande virtuelle pour obtenir le signal décodé vocalement d'extension de largeur de bande correspondant au premier signal décodé vocalement.


 
10. Dispositif de réseau (500) selon la revendication 9, dans lequel :
dans un processus d'estimation, en utilisant le paramètre de décodage vocal, de l'enveloppe spectrale d'extension de largeur de bande correspondant au premier signal décodé vocalement, le premier processeur d'extension de largeur de bande virtuelle (530) est spécialement configuré pour estimer, en utilisant le paramètre de décodage vocal et en se basant sur un procédé de mappage linéaire, sur un procédé de mappage de livre de codes ou sur un procédé de mappage de statistiques, l'enveloppe spectrale d'extension de largeur de bande correspondant au premier signal décodé vocalement.
 
11. Dispositif de réseau (500) selon l'une quelconque des revendications 9 et 10, dans lequel le paramètre de décodage vocal comprend une période de hauteur tonale, un facteur de voisement et un paramètre de codage prédictif linéaire.
 
12. Dispositif de réseau (500) selon l'une quelconque des revendications 9 à 11, dans lequel le dispositif de réseau comprend de multiples décodeurs vocaux et le premier décodeur vocal (520) est un décodeur vocal qui se trouve dans de multiples décodeurs vocaux et qui correspond à la largeur de bande de fréquence maximale prise en charge par le premier terminal ; ou
le dispositif de réseau comprend de multiples processeurs d'extension de largeur de bande virtuelle et le premier processeur d'extension de largeur de bande virtuelle (530) est un processeur d'extension de largeur de bande virtuelle qui se trouve dans de multiples processeurs d'extension de largeur de bande virtuelle et qui correspond à la largeur de bande de fréquence maximale prise en charge par le second terminal ; ou
le dispositif de réseau comprend de multiples codeurs vocaux et le premier codeur vocal (540) est un codeur vocal qui se trouve dans de multiples codeurs vocaux et qui correspond à la largeur de bande de fréquence maximale prise en charge par le second terminal.
 
13. Dispositif de réseau (500) selon l'une quelconque des revendications 9 à 12, dans lequel des largeurs de bande de fréquence du premier signal codé vocalement et du deuxième signal codé vocalement sont deux des largeurs de bande de fréquence suivantes : à bande étroite, à large bande, à super large bande ou à bande complète.
 
14. Dispositif de réseau (500) selon l'une quelconque des revendications 9 à 13, dans lequel le dispositif de réseau comprend en outre un second décodeur vocal (550), un second codeur vocal (570) et un premier dispositif de sous-échantillonnage (560) ; l'interface de communication (510) est en outre configurée pour recevoir un troisième signal codé vocalement en provenance du second terminal ;
le second décodeur vocal (550) est configuré pour effectuer un traitement de décodage vocal sur le troisième signal codé vocalement pour obtenir un troisième signal décodé vocalement ;
le premier dispositif de sous-échantillonnage (560) est configuré pour effectuer un traitement de sous-échantillonnage sur le troisième signal décodé vocalement pour obtenir un quatrième signal décodé vocalement ;
le second codeur vocal (570) est configuré pour effectuer un traitement de codage vocal sur le quatrième signal décodé vocalement pour obtenir un quatrième signal codé vocalement, dans lequel une largeur de bande de fréquence du quatrième signal codé vocalement est inférieure à une largeur de bande de fréquence du troisième signal codé vocalement et un taux d'échantillonnage du quatrième signal codé vocalement est inférieur à un taux d'échantillonnage du troisième signal codé vocalement ; et
l'interface de communication (510) est en outre configurée pour envoyer le quatrième signal codé vocalement au premier terminal ; ou l'interface de communication est en outre configurée : après la réalisation d'un traitement d'amélioration de la voix sur le quatrième signal codé vocalement pour obtenir un quatrième signal codé vocalement obtenu après le traitement d'amélioration de la voix, pour envoyer le quatrième signal codé vocalement obtenu après le traitement d'amélioration de la voix au premier terminal.
 
15. Dispositif de réseau (500) selon la revendication 14, dans lequel le dispositif de réseau comprend de multiples décodeurs vocaux et le second décodeur vocal (550) est un décodeur vocal correspondant à la largeur de bande de fréquence maximale prise en charge par le second terminal ; ou
le dispositif de réseau comprend de multiples dispositifs de sous-échantillonnage et le premier dispositif de sous-échantillonnage (560) est un dispositif de sous-échantillonnage qui se trouve dans de multiples dispositifs de sous-échantillonnage et qui correspond à la largeur de bande de fréquence maximale prise en charge par le premier terminal ; ou
le dispositif de réseau comprend de multiples codeurs vocaux et le second codeur vocal (570) est un codeur vocal correspondant à la largeur de bande de fréquence maximale prise en charge par le premier terminal.
 
16. Dispositif de réseau (500) selon l'une quelconque des revendications 9 à 15, dans lequel le dispositif de réseau est une station de base, un dispositif de commande de réseau radio ou un dispositif de réseau central.
 
17. Système de communication comprenant le dispositif de réseau (500) selon l'une quelconque des revendications 9 à 16.
 




Drawing



































Cited references

REFERENCES CITED IN THE DESCRIPTION



This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

Patent documents cited in the description




Non-patent literature cited in the description