(19)
(11)EP 3 149 730 B1

(12)EUROPEAN PATENT SPECIFICATION

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

(21)Application number: 15727222.0

(22)Date of filing:  22.05.2015
(51)International Patent Classification (IPC): 
G10L 21/0364(2013.01)
G10L 21/0232(2013.01)
(86)International application number:
PCT/US2015/032147
(87)International publication number:
WO 2015/183728 (03.12.2015 Gazette  2015/48)

(54)

ENHANCING INTELLIGIBILITY OF SPEECH CONTENT IN AN AUDIO SIGNAL

VERBESSERUNG DER VERSTÄNDLICHKEIT VON SPRACHINHALTEN IN EINEM AUDIOSIGNAL

AMÉLIORATION DE L'INTELLIGIBILITÉ DU CONTENU PARLÉ D'UN SIGNAL AUDIO


(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: 26.05.2014 CN 201410236155
18.06.2014 US 201462013950 P

(43)Date of publication of application:
05.04.2017 Bulletin 2017/14

(73)Proprietor: Dolby Laboratories Licensing Corporation
San Francisco, CA 94103 (US)

(72)Inventors:
  • MA, Guilin
    Chaoyang District, Beijing 100021 (CN)
  • ZHENG, Xiguang
    Chaoyang District, Beijing 100021 (CN)
  • BROWN, C. Phillip
    San Francisco, California 94103 (US)

(74)Representative: Dolby International AB Patent Group Europe 
Apollo Building, 3E Herikerbergweg 1-35
1101 CN Amsterdam Zuidoost
1101 CN Amsterdam Zuidoost (NL)


(56)References cited: : 
US-A1- 2009 304 215
US-A1- 2012 123 770
US-A1- 2011 054 887
US-B2- 7 010 133
  
  • WARD DOMINIC ET AL: "Multitrack Mixing Using a Model of Loudness and Partial Loudness", AES CONVENTION 133; 20121001, AES, 60 EAST 42ND STREET, ROOM 2520 NEW YORK 10165-2520, USA, 25 October 2012 (2012-10-25), XP040574745,
  • MOORE B C J ET AL: "A MODEL FOR THE PREDICTION OF THRESHOLDS, LOUDNESS, AND PARTIAL LOUDNESS", JOURNAL OF THE AUDIO ENGINEERING SOCIETY, AUDIO ENGINEERING SOCIETY, NEW YORK, NY, US, vol. 45, no. 4, 1 April 1997 (1997-04-01), pages 224-240, XP000700661, ISSN: 1549-4950
  • CHOI J-H ET AL: "Speech Reinforcement Based on Soft Decision under Far-End Noise Environments", IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS,COMMUNICATIONS AND COMPUTER SCIENCES, ENGINEERING SCIENCES SOCIETY, TOKYO, JP, vol. E92A, no. 8, 1 August 2009 (2009-08-01), pages 2116-2119, XP001548396, ISSN: 0916-8508, DOI: 10.1587/TRANSFUN.E92.A.2116
  
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

TECHNOLOGY



[0001] Embodiments of the present application generally relate to signal processing, and more specifically, to enhancing intelligibility of speech content in an audio signal.

BACKGROUND



[0002] Audio signals may contain both speech and non-speech components. The speech component contains speech content while the non-speech component may contain, for example, audio contents in the surround channels of a multichannel audio signal. Furthermore, when the audio signal is played back to users, an environmental noise signal may be simultaneously present external to the audio signal. In order to improve user's experiences, it would be desirable to enhance the intelligibility of the speech content contained in the speech component in the presence of interfering sound signals, such as the non-speech component in the audio signal and/or the environmental noise signal external to the audio signal.

[0003] As used herein, the term "intelligibility of speech content" refers to an indication of the degree of comprehensibility of the speech content. The term "loudness" refers to a perceptual magnitude corresponding to physical strength of the audio signal. The term "partial loudness" refers to the perceived loudness of the audio signal in the presence of interfering sound signals, such as environmental noise signals. The term "environmental noise signal" refers to a noise signal in an ambient environment external to the audio signal. The term "speech component" refers to a component containing speech content in the audio signal, and the term "non-speech component" refers to a component containing non-speech content in the audio signal.

[0004] Some conventional approaches to enhance the intelligibility of the speech content work on the basis of loudness domain processing. In such an approach, the intelligibility of the speech content may be enhanced by controlling partial loudness of the speech component in the audio signal. More specifically, the partial loudness of the speech component is maintained at a reference level of loudness, without taking environmental noise into account. However, there is no mechanism for verifying whether the resulting intelligibility of the speech content is desirable or comfortable to individual users.

[0005] It is also known to enhance the intelligibility of the speech content based on excitation domain processing. The intelligibility of the speech content is enhanced by adjusting the audio signal based on the ratio between a speech component and interfering sound signals. Such approach is applicable in scenarios where the internal interfering sound signal is present or where the external interfering sound signal is present. However, this approach does not work when both the non-speech component and the environmental noise signal are present.
CHOI J-H et al, "Speech Reinforcement Based on Soft Decision under Far-End Noise Environments", discloses estimating and amplifying only the speech signal by considering the speech signal in relation to the near-end noise, thereby avoiding amplification of the far-end noise.
US 7,010,133 discloses a method for automatic amplification adjustment in a hearing aid device.
US 2011/0054887 discloses a method for maintaining speech audibility in multi-channel audio with minimal impact on surround experience. According to the method, first and second characteristics of the multi-channel audio signal are compared to generate an attenuation factor.

SUMMARY



[0006] In order to address the foregoing and other potential problems, the present invention proposes methods and systems for enhancing intelligibility of speech content in an audio signal.

[0007] In one aspect, embodiments of the present invention provide a method for enhancing intelligibility of speech content in an audio signal, the audio signal containing a speech component and a non-speech component, the speech component containing the speech content. The method comprises: obtaining reference loudness of the audio signal, wherein the reference loudness of the audio signal is the loudness of the audio signal without an environmental noise signal; adjusting partial loudness of the audio signal to the reference loudness; calculating a first metric as a signal-to-noise ratio of the speech component to the non-speech component; obtaining a second metric which is a reference signal-to-noise ratio of the speech component to the non-speech component and the environmental noise signal; deriving a boosting gain for boosting the speech component and/or an attenuating gain for attenuating the non-speech component based on a smallest one of the first and the second metric; and enhancing the intelligibility of the speech component by adjusting a ratio of the speech component to the non-speech component and the environmental noise signal based on the first and second metrics. Embodiments in this regard further comprise a corresponding computer program product.

[0008] In another aspect, embodiments of the present invention provide a system for enhancing intelligibility of speech content in an audio signal, the audio signal containing a speech component and a non-speech component, the speech component containing the speech content. The system comprising: a reference loudness obtaining unit configured to obtain reference loudness of the audio signal wherein the reference loudness of the audio signal is the loudness of the audio signal without an environmental noise signal; a loudness adjusting unit configured to adjust partial loudness of the audio signal to the reference loudness of the audio signal; a first metric calculating unit configured to calculate a first metric as a signal-to-noise ratio of the speech component to the non-speech component; a second metric obtaining unit configured to obtain a second metric which is a reference signal-to-noise ratio of the speech component to the non-speech component and an environmental noise signal; a gain determining unit configured to derive a boosting gain for boosting the speech component and/or an attenuating gain for attenuating the non-speech component based on a smallest one of the first and the second metric; and an intelligibility enhancing unit configured to enhance the intelligibility of the speech component by adjusting a ratio of the speech component to the non-speech component and the environmental noise signal based on the first and second metrics.

[0009] Through the following description, it would be appreciated that according to embodiments of one aspect of the present invention, the partial loudness of the audio signal is adjusted based on a degree of the intelligibility of the speech content contained in the speech component of the audio signal such that the intelligibility of the speech content may be enhanced to achieve a certain level of intelligibility. In this way, the intelligibility of the speech content resulted from partial loudness processing may be verified and therefore the high degree of intelligibility may be ensured.

[0010] It would also be appreciated that according to embodiments of another aspect of the present invention, the audio signal is adjusted in the excitation domain based on a ratio of the speech component to the non-speech component and a reference ratio of the speech component to the non-speech component and an environmental noise signal when both the non-speech component and the environmental noise signal are present. In this way, there is provided in the excitation domain a solution directed to the scenario where both the non-speech component and the environmental noise signal are present.

[0011] Other advantages achieved by embodiments of the present invention will become apparent through the following descriptions.

DESCRIPTION OF DRAWINGS



[0012] Through the following detailed description with reference to the accompanying drawings, the above and other objectives, features and advantages of embodiments of the present invention will become more comprehensible. In the drawings, several embodiments of the present invention will be illustrated in an example and non-limiting manner, wherein:

Figure 1 is an example graph illustrating the influence of the environmental noise signal on gains for the audio signal in the partial loudness domain processing;

Figure 2 illustrates a flowchart of an exemplary method for enhancing the intelligibility of speech content in an audio signal;

Figure 3 illustrates a flowchart of another exemplary method for enhancing intelligibility of speech content in an audio signal;

Figure 4 illustrates a flowchart of an exemplary method for determining the target loudness in response to the intelligibility criterion being not met;

Figure 5 is a graph illustrating example relationship between loudness and the ratio of the speech component to the non-speech component and ratio of the speech component to the non-speech component and the environmental noise signal;

Figure 6 illustrates a block diagram of an exemplary system for enhancing the intelligibility of speech content in an audio signal;

Figure 7 illustrates a flowchart of a method for enhancing the intelligibility of speech content in an audio signal according to some example embodiments of the present invention;

Figure 8 is a graph illustrating an example of the frequency dependent metric indicating the reference ratio of the speech component to the non-speech component and the environmental noise signal according to an example embodiment of the present invention;

Figure 9 illustrates a block diagram of a system for enhancing the intelligibility of speech content in an audio signal according to some example embodiments of the present invention; and

Figure 10 illustrates a block diagram of an example computer system suitable for implementing embodiments of the present invention



[0013] Throughout the drawings, the same or corresponding reference symbols refer to the same or corresponding parts.

DESCRIPTION OF EXAMPLE EMBODIMENTS



[0014] Principles of the present invention will now be described with reference to various example embodiments illustrated in the drawings. It should be appreciated that depiction of these embodiments is only to enable those skilled in the art to better understand and further implement the present invention, not intended for limiting the scope of the present invention in any manner.

[0015] As described above, an example approach for enhancing the intelligibility of the speech content in the loudness domain is maintaining the partial loudness of the audio signal at a level of reference loudness without the environmental noise signal. Accordingly, an appropriate gain for modifying the audio signal can be derived to ensure the constant partial loudness of the audio signal in the presence of the environmental noise signal. For example, the loudness of the audio signal without the noise signal is first derived, which is served as the target loudness. Then the appropriate gains for the audio signal are derived for adjusting the partial loudness to the target loudness.

[0016] Generally, the partial loudness of the audio signal decreases with the increase of the loudness of the other interfering sound signals. Thus, the higher the level of the environmental noise signal is, the more gain may be applied to the audio signal.

[0017] Figure 1 is an example graph illustrating the influence of the environmental noise signal on gains for the audio signal in the partial loudness domain processing, wherein the horizontal axis represents the excitation level for the audio signal. As illustrated in Figure 1, the left curve represents the partial loudness under the environmental noise signal of 10 dB, while the right curve represents the partial loudness under the environmental noise signal of 40 dB. In order to maintain the same partial loudness (e.g., 0.1 sone in dB as illustrated in the vertical axis), when the level of the noise signal has been increased from 10 dB to 40 dB, there is required an additional gain of more than 20 dB as illustrated in Figure 1. Thus, by applying the appropriate gains, the partial loudness of the audio signal can be preserved under different levels of noise signals. As described above, there is no mechanism for verifying whether the resulting intelligibility of the speech content is desirable in the conventional approach.

[0018] In the following will be discussed examples of a method and system for enhancing the intelligibility of the speech content such that the enhanced intelligibility achieves a certain degree of intelligibility, for example, meets a certain intelligibility criterion. After the partial loudness of the speech content is adjusted to reference loudness, e.g., the loudness without the environmental noise signal, it is determined whether the resulting intelligibility achieves a certain degree of intelligibility. If the resulting intelligibility does not achieve the certain degree of intelligibility, the partial loudness of the speech content will be further adjusted based on the determination result. In this way, the intelligibility of the speech content resulted from partial loudness processing may be verified and therefore the high degree of intelligibility may be ensured.

[0019] Now reference is made to Figure 2 which illustrates a flowchart of a method 200 for enhancing the intelligibility of speech content in an audio signal.

[0020] The audio signal may include at least a speech component which contains the speech content. Optionally, the audio signal may contain a non-speech component. When the speech component is mixed with the non-speech component in the audio signal, the speech and non-speech components may be separated by applying, for example, a technique of blind source separation. Alternatively, the speech and non-speech components may be separated directly when object-based audio format is employed, wherein it is known in advance whether the center channel of a multichannel audio signal contains speech or non-speech object tracks.

[0021] The method 200 may be applied to the following three scenarios: 1) a speech component and an environmental noise signal are present; 2) a speech component and a non-speech component are present; 3) a speech component, a non-speech component and an environmental noise signal are present. Now the method 200 will be described in detail with respect to Figure 2.

[0022] As shown in Figure 2, at step S201, a reference loudness of the audio signal is obtained. Then, at step S202, the partial loudness of the audio signal is adjusted based on the reference loudness and a degree of intelligibility of the speech content such that the intelligibility of the speech content may be enhanced. The degree of the intelligibility of the speech content may be represented by a value, e.g., a score of the intelligibility. Alternatively or additionally, the degree of the intelligibility may be represented by a level from a group consisting of several predefined levels such as high, medium, low, and the like.

[0023] With the method 200, the partial loudness of the audio signal is not necessarily always fixed at a level of specific reference loudness. Instead, the partial loudness of the audio signal may be adjusted dynamically based on the degree of the intelligibility of the speech content.

[0024] The method 200 may be iteratively performed until the desirable degree of the intelligibility of the speech content is achieved, which will be described below in detail with respect to Figure 2.

[0025] When the method 200 is performed initially, at step S201, the initial reference loudness may be set as the loudness of the audio signal without interfering sound signals. Specifically, in a scenario where a speech component and an environmental noise signal are present, the initial reference loudness may be set as the loudness of the speech component without the environmental noise signal. In another scenario where a speech component and a non-speech component are present, the initial reference loudness may be set as the loudness of the speech component without the non-speech component. In yet another scenario where a speech component, a non-speech component and an environmental noise signal are present, the initial reference loudness maybe set as the loudness of the speech component without the non-speech component and the environmental noise signal.

[0026] Then, at step S202, the partial loudness of the audio signal is adjusted based on the initial reference loudness and the achieved degree of the intelligibility after the use of the initial reference loudness in adjusting the partial loudness. If the currently achieved degree of the intelligibility of the speech content is undesirable, the reference loudness is increased by an increment, and the method 200 is iterated until the desirable degree of the intelligibility of the speech content is achieved.

[0027] Alternatively, the method 200 may be performed only once and the partial loudness of the audio signal is adjusted to an appropriate loudness. The appropriate loudness may be determined according to the initial reference loudness and the desirable degree of the intelligibility.

[0028] For the implementation of adjusting the partial loudness of the audio signal, the partial loudness of the speech component may be increased so as to enhance the intelligibility of the speech content. Specifically, at step S202, the partial loudness of the speech component may be increased based on the reference loudness and the degree of the intelligibility of the speech content such that the intelligibility of the speech content may be enhanced.

[0029] Alternatively, if the audio signal also contains a non-speech component, the partial loudness of the non-speech component may be reduced so as to enhance the intelligibility of the speech content. Specifically, at step S202, the partial loudness of the non-speech component may be reduced based on the reference loudness and the degree of the intelligibility of the speech content such that the intelligibility of the speech content may be enhanced.

[0030] Alternatively, at step S202, the partial loudness of the speech component may be increased and the partial loudness of the non-speech component may be reduced at the same time. It would be appreciated that in the case where the partial loudness of the non-speech component is adjusted, the reference loudness related to the non-speech component may be obtained. With the adjustment of the non-speech component, the level of the speech component may not need to be changed a lot, and thereby the change of timbre of the speech content may be reduced.

[0031] Figure 3 illustrates a flowchart of a method 300 for enhancing intelligibility of speech content in an audio signal. The method 300 may be implemented after the reference loudness of the audio signal is obtained, for example, in the method 200.

[0032] In the method 300, an intelligibility criterion is used for determining the degree of the intelligibility of the speech content such that an evaluation of the degree of the intelligibility may be introduced to ensure the high degree of the intelligibility of the speech content resulted from the partial loudness processing.

[0033] As illustrated in Figure 3, in the method 300, at step S301, the partial loudness of the audio signal is adjusted to the reference loudness after the reference loudness is obtained, for example, at step S201 of the method 200. In this way, the intelligibility of the speech content may achieve a certain degree of the intelligibility.

[0034] Next, at step S302, it is determined whether an intelligibility criterion is met by the intelligibility of the speech content in the adjusted audio signal. As such, an evaluation of the achieved degree of the intelligibility of the speech content after the previous partial loudness processing may be introduced.

[0035] In order to evaluate the intelligibility of the speech content based on the intelligibility criterion, a score of the intelligibility of the speech content may be calculated, wherein more score indicates the higher degree of the intelligibility of the speech content. It should be noted that any other approach of the evaluation of the intelligibility of the speech content may be employed.

[0036] After the step of the determination in the method 300, if the criterion is met, it means that the currently achieved intelligibility of the speech content is desirable. Thus, there is no need for additional loudness for adjusting the partial loudness of the audio signal, and the method 300 ends.

[0037] If the criterion is not met, it means the currently achieved intelligibility of the speech content is undesirable. Then, the method proceeds to step S303, where target loudness is determined in response to the intelligibility criterion being not met. Then, at step S304, the partial loudness of the audio signal is adjusted to the target loudness. As such, the intelligibility of the speech content may be further enhanced with the introduction of the evaluation of the degree of the intelligibility.

[0038] As described with respect to Figure 2, the method 300 in Figure 3 may also be iteratively performed until the desirable degree of the intelligibility of the speech content is achieved; alternatively, the method 300 may be performed only once and the partial loudness of the audio signal may be accordingly adjusted to the appropriate loudness for achieving the desirable degree of intelligibility of the speech content.

[0039] Specifically, the target loudness may be determined iteratively. For example, whenever the intelligibility criterion is not met, the target loudness is increased by an increment, e.g., minimum amount of the loudness. Then, the partial loudness of the audio signal may be adjusted based on the new target loudness. Next, it is determined again whether the enhanced intelligibility of the speech content meets the intelligibility criterion. The method is iterated until the intelligibility criterion is met.

[0040] The target loudness may be determined once based on the degree of the intelligibility of the speech content, e.g., using a mapping function, for example, between the intelligibility and the loudness. The mapping function may be derived from empirical psychoacoustic studies.

[0041] Similar to the examples described with respect to Figure 2, the method 300 may also be applied to the following three scenarios: 1) a speech component and an environmental noise signal are present; 2) a speech component and a non-speech component are present; 3) a speech component, a non-speech component and an environmental noise signal are present.

[0042] Likewise, as described with respect to Figure 2, the intelligibility of the speech content may be enhanced by at least one of increasing the partial loudness of the speech component and reducing the partial loudness of the non-speech component. For the sake of briefness, the detailed description is omitted.

[0043] Figure 4 illustrates a flowchart of a method 400 for determining the target loudness in response to the intelligibility criterion being not met.

[0044] It would be appreciated that the method 400 may be applied to the scenario where a speech component, a non-speech component and an environmental noise signal are present.

[0045] Before the method 400 is performed, the partial loudness of the audio signal may be adjusted to the reference loudness without the environmental noise signal using the above described methods, and the determination whether the intelligibility criterion is met may also be performed using the above described methods.

[0046] In the method 400, the intelligibility of the speech content contained by the speech component may be ensured, while the simultaneously occurring no-speech component may be audible so as to ensure the immersion of the whole audio signal and thereby improve the user's experiences. Now the method 400 will be described in detail with respect to Figure 4.

[0047] In response to the intelligibility criterion being not met by the intelligibility of the speech content, the method 400 starts.

[0048] In the method 400, at step S401, a first metric is calculated for indicating a ratio of the speech component to the non-speech component. Then, at step S402, a second metric is calculated for indicating a ratio of the speech component to the non-speech component and an environmental noise signal. Next, at step S403, additional loudness for adjusting the partial loudness of the audio signal is determined based on the first and second metrics. Then, at step S404, the target loudness is determined based on the reference loudness and the additional loudness.

[0049] The first and second metrics may be any form of metrics which indicate the ratio of the speech component to the non-speech component and the reference ratio of the speech component to the non-speech component and the environmental noise signal, respectively. For example, the metrics may be the logarithm or any other appropriate functions of the ratios.

[0050] It would be appreciated that the difference between the first and second metrics may indicate the interference of the environmental noise signal on the audio signal. With the adjustment of the partial loudness of the audio signal based on the first metric, which indicates a ratio of the speech component to the non-speech component, and the second metric, which indicates a reference ratio of the speech component to the non-speech component and the environmental noise signal, the desirable audio playback quality in the presence of the environmental noise signal may be ensured.

[0051] At steps S401 and S402, the first and second metrics may be calculated at least partially based on a frequency band of the audio signal. It is known that the contributions of different frequency bands to the intelligibility of the speech content may be different. With the above process of calculation, the intelligibility of the speech content may be further enhanced.

[0052] Before the step S402 of the method 400, the partial loudness of the audio signal containing the speech and non-speech components is first adjusted to the reference loudness without the presence of the environmental noise signal using the above described methods. Thus, the loudness of audio signal is enhanced so that the whole audio playback quality may be ensured.

[0053] Specifically, the first and second metrics are both calculated and weighted for a frequency band of the audio signal. The calculated first metric is given by the following Equations (1):

where SARSI represents the first metric, b represents a frequency band of the audio signal, W(b) represents the weight value for a frequency band, b, Ss (b) represents the speech component of the audio signal for a frequency band, b, Sns (b) represents the non-speech component of the audio signal for a frequency band, b, Tmax represents the maximum threshold, and Tmin represents the minimum threshold.

[0054] As described above, the second metric may be calculated after the partial loudness of the audio signal containing the speech and non-speech components is adjusted. In this case, the second metric may be calculated and weighted for each frequency band of the audio signal as given in the following Equations (2):

where SNARSI represents the second metric, b represents a frequency band of the audio signal, W(b) represents the weight value for a frequency band, b, SLR-s (b) represents the partial loudness adjusted speech component of the audio signal for a frequency band, b, SLR-ns (b) represents the partial loudness adjusted non-speech component of the audio signal for a frequency band, b, Next (b) represents the environmental noise signal for a frequency band, b, Tmax represents the maximum threshold, and Tmin represents the minimum threshold.

[0055] W(b) in Equations (1) and (2) is determined based on the impact of the frequency band to the intelligibility of the speech content. For example, W(b) may be higher, if the frequency band, b, has more impact to the intelligibility of the speech content. The weight may be derived from the speech intelligibility studies and standards, such as the Speech Intelligibility Index (SII, see ANSI S3.5-1997, "Methods for Calculation of the Speech Intelligibility Index") and Articulation Index (AI, see Mueller, G. & Killion, M. (1992)., "An Easy Method for Calculating the Articulation Index", The Hearing Journal, 45(9), 14-17). W(b) may meet the following condition:



[0056] The thresholds Tmax and Tmin in Equations (1) and (2) may be used for constraining the first and second metrics within a certain range, e.g., suitable for human's perception such that extremely high or low physical strength of the audio signal is avoided, thereby improving user's experiences. It should be noted that no use of the thresholds may also be feasible.

[0057] At step S403, the additional loudness for adjusting the partial loudness of the audio signal is determined based on the difference between the first and second metrics.

[0058] Example relationship between the difference of SARSII and SNARSII and the additional loudness (AL) is illustrated in Figure 5. As illustrated in Figure 5, AL is increased with the increase of the difference between SARSII and SNARSII, wherein SARSII and SNARSII are determined based on the standard of SII.

[0059] Alternatively, the additional loudness may be derived by a defined SNARSI to additional loudness mapping function, which may be derived from empirical psychoacoustic studies. Alternatively, the mapping function may be derived by recording user behavior to determine the mapping function adaptively.

[0060] After the additional loudness, AL, is determined, the target loudness is given by the following Equation (4):

where L0 represents the reference loudness.

[0061] It should be noted the calculation of the first and second metrics, and the determination of the additional loudness and the target loudness as discussed above are just for the purpose of illustration.

[0062] As described with respect to Figures 2 and 3, the partial loudness of both the speech and non-speech components may be adjusted. After step S404 of the method 400, the appropriate gain to be applied to the speech component may be derived for each frequency band such that the partial loudness of the speech component is adjusted to the target loudness. Alternatively, the appropriate gain to be applied to the non-speech component may be derived for each frequency band such that the non-speech component may be adjusted to the target loudness.

[0063] Figure 6 illustrates a block diagram of a system 600 for enhancing the intelligibility of speech content in an audio signal.

[0064] As illustrated in Figure 6, the system 600 may comprise a reference obtaining unit 601 and an intelligibility enhancing unit 602. The reference loudness obtaining unit 601 may be configured to obtain reference loudness of the audio signal. The intelligibility enhancing unit 602 may be configured to enhance the intelligibility of the speech content by adjusting partial loudness of the audio signal based on the reference loudness and a degree of the intelligibility.

[0065] The intelligibility enhancing unit 602 may comprise a loudness adjusting unit configured to increase the partial loudness of the speech component based on the reference loudness and the degree of the intelligibility.

[0066] Optionally, the intelligibility enhancing unit 602 may comprise a loudness adjusting unit configured to reduce the partial loudness of the non-speech component based on the reference loudness and the degree of the intelligibility in response to a determination that the audio signal contains a non-speech component.

[0067] The intelligibility enhancing unit 602 may comprise a loudness adjusting unit configured to adjust the partial loudness of the audio signal to the reference loudness and adjust the partial loudness of the audio signal to a target loudness in response to an intelligibility criterion being not met; an intelligibility determining unit configured to determine whether the intelligibility criterion is met by the intelligibility of the speech content in the adjusted audio signal; a target loudness determining unit configured to determine the target loudness in response to the intelligibility criterion being not met.

[0068] The target loudness determining unit may comprise a first metric calculating unit configured to calculate a first metric indicating a ratio of the speech component to the non-speech component; a second metric calculating unit configured to calculate a second metric indicating a ratio of the speech component to the non-speech component and an environmental noise signal; an additional loudness determining unit configured to determine additional loudness based on the first and second metrics; and a determining unit configured to determine the target loudness based on the reference loudness and the additional loudness.

[0069] Additionally, the first metric calculating unit may be further configured to calculate the first metric at least partially based on a frequency band of the audio signal. The second metric calculating unit may be further configured to calculate the second metric at least partially based on the frequency band of the audio signal.

[0070] For the sake of clarity, some optional components of the system 600 are not illustrated in Figure 6. However, it should be appreciated that the features as described above with reference to Figures 2-4 are all applicable to the system 600. Moreover, the components of the system 600 may be a hardware module or a software unit module. For example, the system 600 may be implemented partially or completely with software and/or firmware, for example, implemented as a computer program product embodied in a computer readable medium. Alternatively or additionally, the system 600 may be implemented partially or completely based on hardware, for example, as an integrated circuit (IC), an application-specific integrated circuit (ASIC), a system on chip (SOC), a field programmable gate array (FPGA), and so forth.

[0071] With respect to Figures 2-6, a method and system for enhancing the intelligibility of the speech content have been described above, which may enable the enhanced intelligibility to achieve a certain level of intelligibility by introducing the evaluation of degree of the intelligibility of the speech content in adjusting the partial loudness of the speech component.

[0072] As described above, in the excitation domain, an example approach for enhancing the intelligibility of the speech content is aimed at boosting the speech component relative to either the non-speech component or the environmental noise signal. In the excitation domain processing, there is no solution directed to the scenario where both the non-speech component and the environmental noise signal are present.

[0073] According to an aspect of the present invention, in order to address the above and other potential problems, some embodiments of the present invention proposes a method and system for enhancing the intelligibility of the speech content by adjusting the audio signal in the excitation domain when both the non-speech component and the environmental noise signal are present.

[0074] Now reference is made to Figure 7 which illustrates a flowchart of a method 700 for enhancing the intelligibility of speech content in an audio signal according to some example embodiments of the present invention.

[0075] In the embodiments of the present invention, the audio signal may contain both a speech component and a non-speech component. As described with respect to Figure 2, the speech and non-speech components may be separated by applying, for example, a technique of blind source separation, or, alternatively, separated directly when object-based audio format is employed. Furthermore, an environmental noise signal may be simultaneously present external to the audio signal.

[0076] As illustrated in Figure 7, in the method 700, at step S701, a first metric is calculated for indicating a ratio of the speech component to the non-speech component. Then, at step S702, a second metric is obtained for indicating a reference ratio of the speech component to the non-speech component and the environmental noise signal. Next, at step S703, the intelligibility of the speech component is enhanced by adjusting a ratio of the speech component to the non-speech component and the environmental noise signal based on the first and second metrics.

[0077] With the method 700, the solution for enhancing the intelligibility of the speech content is provided in the excitation domain in the scenario where the environmental noise signal is simultaneously present external the audio signal.

[0078] In an embodiment of the present invention, at step S703 of the method 700, the first and second metrics may be compared. If the first metric is less than the second metric, the ratio of the speech component to the non-speech component is adjusted to the first metric, or, otherwise, adjusted to the second metric. As such, less timbre change of the speech signal may be the result from the enhancement of intelligibility of the speech content. It should be noted that the specific approach for adjusting the ratio of the speech component to the non-speech component and the environmental noise signal based on the first and second metrics is not limited to the determination of the lesser one of the first and second metrics as a target of the adjustment discussed above, which is only for the purpose of illustration, but not for the purpose of limitation of the scope of the present invention.

[0079] Optionally, in an embodiment of the present invention, before the first metric indicating the ratio of the speech component to the non-speech component is calculated, reference loudness of the audio signal may be obtained. Then, partial loudness of the audio signal may be adjusted to the reference loudness of the audio signal. In an example embodiment of the present invention, the reference loudness may be the loudness of the audio signal without the environmental noise signal. It should be noted that other reference loudness may be employed instead, and the scope of the invention may not be limited in this regard. After such a pre-processing stage, both the speech component and the non-speech component may be enabled to be heard by the users when the environmental noise signal is present, thereby ensuring the immersion of the whole audio signal.

[0080] Optionally, in an embodiment of the present invention, at step S703 of the method 700, the ratio of the speech component to the non-speech component and the environmental noise signal is adjusted during a speech section, which contains at least a part of the speech component, and thereby the efficiency of the adjustment may be ensured.

[0081] As described above with respect to Figure 4, the contributions of different frequency bands to the intelligibility of the speech content may be different. The method 700 as illustrated in Figure 7 may be performed based on each frequency band of the audio signal according to some embodiments of the present invention, which will be described below in detail with respect to Figure 7.

[0082] In an embodiment of the present invention, at step S701 of the method 700, the first metric indicating the ratio of the speech component to the non-speech component may be calculated for a frequency band of the audio signal, specifically, the calculated first metric for a frequency band is given by the following Equation (5):

where b represents a frequency band of the audio signal, SAR(b) represents the first metric for a frequency band, b, Ss (b) represents the speech component of the audio signal for a frequency band, b, and Sns (b) represents the non-speech component of the audio signal for a frequency band, b.

[0083] Next, at step S702, the second metric indicating the reference ratio of the speech component to the non-speech component and the environmental noise signal may be obtained at least partially based on the frequency band. For example, the second metric may be derived from the speech intelligibility studies and standards, such as the Speech Intelligibility Index (SII) and Articulation Index (AI), as described above.

[0084] Figure 8 illustrates an example of the frequency dependent metric indicating the reference ratio of the speech component to the non-speech component and the environmental noise signal according to an example embodiment of the present invention. As illustrated in Figure 8, the metric, which is represented by reference SNR in Figure 8, for the frequency bands of higher importance are larger. It should be noted that the above metrics are only for the purpose of illustration, any frequency dependent metric that reflects the importance of the frequency bands may be employed, and the scope of the invention should not be limited in this regard.

[0085] Then, at step S703, the first metric and the second metric may first be compared. Then, the lesser one of the two metrics may be determined as an adjusting target, as given by the following Equation (6):

where b represents a frequency band of the audio signal, SAR(b) represents the first metric for a frequency band, b, and refSNR(b) represents the second metric for a frequency band, b.

[0086] After the adjusting target is determined, the ratio of the speech component to the non-speech component and the environmental noise signal may be adjusted based on the adjusting target.

[0087] In some embodiments of the present invention, at step S703 of the method 700, the adjustment of the ratio of the speech component to the non-speech component and the environmental noise signal may be achieved by boosting the speech component, or, alternatively, by attenuating the non-speech component.

[0088] Specifically, in an embodiment of the present invention, once the adjusting target has been determined, a boosting gain g to be applied to the speech component may be derived from the following Equation (7):



[0089] Alternatively, in another embodiment of the present invention, an attenuating gain g to be applied to the non-speech component may be derived from the following Equation (8):

where the following condition may be met:



[0090] Alternatively, in yet another embodiment of the present invention, both the boosting gain for the speech component and the attenuation gain for the non-speech component may be derived.

[0091] It should be noted the determination of the first and second metrics, the adjusting target and adjusting gains as discussed above are just for the purpose of illustration, without limiting the scope of the present invention. It would be appreciated that, the first and second metrics may be any form of metrics which indicate the ratio of the speech component to the non-speech component and the ratio of the speech component to the non-speech component and the environmental noise signal, respectively. For example, the metrics may be the logarithm or any other appropriate functions of the ratios. The scope of the present invention should not be limited in this regard.

[0092] Alternatively, in order to derive appropriate gains for the speech and/or non-speech component, in an embodiment of the present invention, an iterative search may be performed among the candidate gain(s) such that a certain criterion is met. An example criterion may be that the desirable degree of the intelligibility of the speech content is achieved, while minimum modification gains are applied to the audio signal.

[0093] In an embodiment of the present invention, after the gains are derived, it may be further constrained, for example, by employing some compression curves such that, for example, less gain would be applied when the loudness of the external noise is low and vice versa. As such, the derived gains may be further smoothed to avoid sudden change of audio timbre and/or signal power.

[0094] Figure 9 illustrates a block diagram of a system 900 for enhancing the intelligibility of speech content in an audio signal according to some example embodiments of the present invention.

[0095] As illustrated in Figure 9, the system 900 comprises a first metric calculating unit 901, a second metric obtaining unit 902 and an intelligibility enhancing unit 903. The first metric calculating unit 901 maybe configured to calculate a first metric indicating a ratio of the speech component to the non-speech component. The second metric obtaining unit 902 may be configured to obtain a second metric indicating a reference ratio of the speech component to the non-speech component and an environmental noise signal. The intelligibility enhancing unit 903 may be configured to enhance the intelligibility of the speech component by adjusting a ratio of the speech component to the non-speech component and the environmental noise signal based on the first and second metrics.

[0096] In some embodiments of the present invention, the intelligibility enhancing unit 903 may comprise a comparing unit configured to compare the first and second metrics; a ratio adjusting unit configured to adjust the ratio based on the first metric in response to the first metric being less than the second metric and adjust the ratio based on the second metric in response to the first metric being larger than the second metric.

[0097] In some embodiments of the present invention, the system 900 may further comprise a reference loudness obtaining unit configured to obtain reference loudness of the audio signal; and a loudness adjusting unit configured to adjust partial loudness of the audio signal to the reference loudness of the audio signal. In the embodiments of the present invention, the first metric calculating unit may be configured to calculate the first metric based on the adjusted audio signal.

[0098] In some embodiments of the present invention, the intelligibility enhancing unit 903 may comprise a gain determining unit configured to determine a gain to be applied to the audio signal based on the first and second metrics; a gain constraining unit configured to constrain the determined gain based on the loudness of the environmental noise signal; and a gain applying unit configured to apply the constrained gain to the audio signal.

[0099] For the sake of clarity, some optional components of the system 900 are not illustrated in Figure 9. However, it should be appreciated that the features as described above with reference to Figures 7 and 8 are all applicable to the system 900. Moreover, the components of the system 900 may be a hardware module or a software unit module. For example, in some embodiments of the present invention, the system 900 maybe implemented partially or completely with software and/or firmware, for example, implemented as a computer program product embodied in a computer readable medium. Alternatively or additionally, the system 900 may be implemented partially or completely based on hardware, for example, as an integrated circuit (IC), an application-specific integrated circuit (ASIC), a system on chip (SOC), a field programmable gate array (FPGA), and so forth. The scope of the present invention is not limited in this regard.

[0100] Figure 10 illustrates a block diagram of an example computer system 1000 suitable for implementing embodiments of the present invention. As illustrated, the computer system 1000 comprises a central processing unit (CPU) 1001 which is capable of performing various processes according to a program stored in a read only memory (ROM) 1002 or a program loaded from a storage section 1008 to a random access memory (RAM) 1003. In the RAM 1003, data required when the CPU 1001 performs the various processes or the like is also stored as required. The CPU 1001, the ROM 1002 and the RAM 1003 are connected to one another via a bus 1004. An input/output (I/O) interface 1005 is also connected to the bus 1004.

[0101] The following components are connected to the I/O interface 1005: an input section 1006 including a keyboard, a mouse, or the like; an output section 1007 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), or the like, and a loudspeaker or the like; the storage section 1008 including a hard disk or the like; and a communication section 1009 including a network interface card such as a LAN card, a modem, or the like. The communication section 1009 performs a communication process via the network such as the internet. A drive 1010 is also connected to the I/O interface 1005 as required. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like, is mounted on the drive 1010 as required, so that a computer program read therefrom is installed into the storage section 1008 as required.

[0102] Specifically, according to embodiments of the present invention, the processes described above with reference to Figures 2-5, 7 and 8 may be implemented as computer software programs. For example, embodiments of the present invention comprise a computer program product including a computer program tangibly embodied on a machine readable medium, the computer program including program code for performing methods 200, 300, 400 and/or 700. In such embodiments, the computer program may be downloaded and mounted from the network via the communication section 1009, and/or installed from the removable medium 1011.

[0103] Generally speaking, various example embodiments of the present invention may be implemented in hardware or special purpose circuits, software, logic or any combination thereof. Some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software which may be executed by a controller, microprocessor or other computing device. While various aspects of the example embodiments of the present invention are illustrated and described as block diagrams, flowcharts, or using some other pictorial representation, it will be appreciated that the blocks, apparatus, systems, techniques or methods described herein may be implemented in, as non-limiting examples, hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof.

[0104] Additionally, various blocks illustrated in the flowcharts may be viewed as method steps, and/or as operations that result from operation of computer program code, and/or as a plurality of coupled logic circuit elements constructed to carry out the associated function(s). For example, embodiments of the present invention include a computer program product comprising a computer program tangibly embodied on a machine readable medium, the computer program containing program codes configured to carry out the methods as described above.

[0105] In the context of the disclosure, a machine readable medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine readable medium may be a machine readable signal medium or a machine readable storage medium. A machine readable medium may include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine readable storage medium would include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0106] Computer program code for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer program codes may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor of the computer or other programmable data processing apparatus, cause the functions/operations specified in the flowcharts and/or block diagrams to be implemented. The program code may execute entirely on a computer, partly on the computer, as a stand-alone software package, partly on the computer and partly on a remote computer or entirely on the remote computer or server.

[0107] Further, while operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order illustrated or in sequential order, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are contained in the above discussions, these should not be construed as limitations on the scope of any invention or of what may be claimed, but rather as descriptions of features that may be specific to particular embodiments of particular inventions. Certain features that are described in this specification in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable sub-combination.

[0108] Various modifications, adaptations to the foregoing example embodiments of this invention may become apparent to those skilled in the relevant arts in view of the foregoing description, when read in conjunction with the accompanying drawings. Furthermore, other embodiments of the inventions set forth herein will come to mind to one skilled in the art to which these embodiments of the invention pertain having the benefit of the teachings presented in the foregoing descriptions and the drawings.

[0109] It will be appreciated that the embodiments of the invention are not to be limited to the specific embodiments disclosed and that modifications and other embodiments are intended to be included within the scope of the appended claims. Although specific terms are used herein, they are used in a generic and descriptive sense only and not for purposes of limitation.


Claims

1. A method for enhancing intelligibility of speech content in an audio signal, the audio signal containing a speech component and a non-speech component, the speech component containing the speech content, the method comprising:

obtaining reference loudness of the audio signal, wherein the reference loudness of the audio signal is the loudness of the audio signal without an environmental noise signal;

adjusting partial loudness of the audio signal to the reference loudness;

calculating a first metric as a signal-to-noise ratio of the speech component to the non-speech component of the adjusted audio signal;

obtaining a second metric which is a reference signal-to-noise ratio of the speech component to the non-speech component and the environmental noise signal as derived from a speech intelligibility standard;

deriving a boosting gain for boosting the speech component and/or an attenuating gain for attenuating the non-speech component based on a smallest one of the first and the second metric; and

enhancing the intelligibility of the speech content by adjusting a ratio of the speech component to the non-speech component and the environmental noise signal, wherein the ratio is adjusted by applying the boosting gain to the speech component and/or applying the attenuating gain to the non-speech component.


 
2. The method according to Claim 1, wherein the ratio of the speech component to the non-speech component and the environmental noise signal is adjusted during a speech section, the speech section containing at least a part of the speech component.
 
3. The method according to Claim 2, wherein the first metric is calculated for a frequency band of the audio signal, and
wherein the second metric is obtained at least partially based on the frequency band.
 
4. The method according to any of Claims 1 to 3, further comprising:
constraining the determined boosting gain and/or attenuating gain based on the loudness of the environmental noise signal prior to applying the boosting gain to the speech component and/or applying the attenuating gain to the non-speech component.
 
5. A system for enhancing intelligibility of speech content in an audio signal, the audio signal containing a speech component and a non-speech component, the speech component containing the speech content, the system comprising:

a reference loudness obtaining unit configured to obtain reference loudness of the audio signal wherein the reference loudness of the audio signal is the loudness of the audio signal without an environmental noise signal;

a loudness adjusting unit configured to adjust partial loudness of the audio signal to the reference loudness of the audio signal;

a first metric calculating unit configured to calculate a first metric as a signal-to-noise ratio of the speech component to the non-speech component of the adjusted audio signal;

a second metric obtaining unit configured to obtain a second metric which is a reference signal-to-noise ratio of the speech component to the non-speech component and the environmental noise signal as derived from a speech intelligibility standard;

a gain determining unit configured to derive a boosting gain for boosting the speech component and/or an attenuating gain for attenuating the non-speech component based on a smallest one of the first and the second metric; and

an intelligibility enhancing unit configured to enhance the intelligibility of the speech component by adjusting a ratio of the speech component to the non-speech component and the environmental noise signal, wherein the ratio is adjusted by applying boosting gain to the speech component and/or applying the attenuating gain to the non-speech component.


 
6. The system according to Claim 5, wherein the ratio adjusting unit is further configured to adjust the ratio of the speech component to the non-speech component and the environmental noise signal during a speech section, the speech section containing at least a part of the speech component.
 
7. The system according to any of Claims 5 to 6, wherein the first metric calculating unit is further configured to calculate the first metric for a frequency band of the audio signal, and
wherein the second metric obtaining unit is further configured to obtain the second metric at least partially based on the frequency band.
 
8. The system according to any of Claims 5 to 7, wherein the intelligibility enhancing unit comprises:
a gain constraining unit configured to constrain the determined boosting gain and/or attenuating gain based on the loudness of the environmental noise signal prior to applying the boosting gain to the speech component and/or applying the attenuating gain to the non-speech component.
 
9. A computer program product for enhancing intelligibility of speech content in an audio signal, the computer program product being tangibly stored on a non-transient computer-readable medium and comprising machine executable instructions which, when executed, cause the machine to perform the method according to any of Claims 1 to 4.
 


Ansprüche

1. Verfahren zur Verbesserung der Verständlichkeit von Sprachinhalten in einem Audiosignal, wobei das Audiosignal eine Sprachkomponente und eine Nicht-Sprachkomponente enthält, wobei die Sprachkomponente den Sprachinhalt enthält, wobei das Verfahren umfasst:

Erhalten der Referenzlautstärke des Audiosignals, wobei die Referenzlautstärke des Audiosignals die Lautstärke des Audiosignals ohne ein Umgebungsrauschsignal ist;

Einstellen der Teillautstärke des Audiosignals auf die Referenzlautstärke;

Berechnen einer ersten Metrik als Signal-Rausch-Verhältnis der Sprachkomponente zur Nicht-Sprachkomponente des eingestellten Audiosignals;

Erhalten einer zweiten Metrik, die ein Referenzsignal-Rausch-Verhältnis der Sprachkomponente zur Nicht-Sprachkomponente und zum Umgebungsrauschsignal, wie von einem Sprachverständlichkeitsstandard abgeleitet, ist;

Ableiten einer Verstärkungssteigerung zum Verstärken der Sprachkomponente und/oder einer Dämpfungssteigerung zum Dämpfen der Nicht-Sprachkomponente basierend auf einer kleinsten der ersten und zweiten Metrik; und

Verbessern der Verständlichkeit des Sprachinhalts durch Einstellen eines Verhältnisses der Sprachkomponente zur Nicht-Sprachkomponente und zum Umgebungsrauschsignal, wobei das Verhältnis durch Anlegen der Verstärkungssteigerung an die Sprachkomponente und/oder Anlegen der Dämpfungssteigerung an die Nicht-Sprachkomponente eingestellt wird.


 
2. Verfahren nach Anspruch 1, wobei das Verhältnis der Sprachkomponente zur Nicht-Sprachkomponente und des Umgebungsrauschsignals während eines Sprachabschnitts eingestellt wird, wobei der Sprachabschnitt mindestens einen Teil der Sprachkomponente enthält.
 
3. Verfahren nach Anspruch 2, wobei die erste Metrik für ein Frequenzband des Audiosignals berechnet wird, und
wobei die zweite Metrik zumindest teilweise basierend auf dem Frequenzband erhalten wird.
 
4. Verfahren nach einem der Ansprüche 1 bis 3, weiter umfassend:
Einschränken der bestimmten Verstärkungssteigerung und/oder Dämpfungssteigerung basierend auf der Lautstärke des Umgebungsrauschsignals vor dem Anlegen der Verstärkungssteigerung an die Sprachkomponente und/oder dem Anlegen der Dämpfungssteigerung an die Nicht-Sprachkomponente.
 
5. System zur Verbesserung der Verständlichkeit von Sprachinhalten in einem Audiosignal, wobei das Audiosignal eine Sprachkomponente und eine Nicht-Sprachkomponente enthält, wobei die Sprachkomponente den Sprachinhalt enthält, wobei das System umfasst:

eine Referenzlautstärke-Erhaltungseinheit, die konfiguriert ist, um die Referenzlautstärke des Audiosignals zu erhalten, wobei die Referenzlautstärke des Audiosignals die Lautstärke des Audiosignals ohne ein Umgebungsrauschsignal ist;

eine Lautstärke-Einstelleinheit, die konfiguriert ist, um die Teillautstärke des Audiosignals an die Referenzlautstärke des Audiosignals anzupassen;

eine Berechnungseinheit für die erste Metrik, die konfiguriert ist, um eine erste Metrik als Signal-Rausch-Verhältnis der Sprachkomponente zur Nicht-Sprachkomponente des eingestellten Audiosignals zu berechnen;

eine Erhaltungseinheit für die zweite Metrik, die konfiguriert ist, um eine zweite Metrik zu erhalten, die ein Referenzsignal-Rausch-Verhältnis der Sprachkomponente zur Nicht-Sprachkomponente und zum Umgebungsrauschsignal, wie von einem Sprachverständlichkeitsstandard abgeleitet, ist;

eine Steigerungsbestimmungseinheit, die konfiguriert ist, um eine Verstärkungssteigerung zum Verstärken der Sprachkomponente und/oder eine Dämpfungssteigerung zum Dämpfen der Nicht-Sprachkomponente basierend auf einer kleinsten der ersten und zweiten Metrik abzuleiten; und

eine Einheit zur Verbesserung der Verständlichkeit, die konfiguriert ist, um die Verständlichkeit der Sprachkomponente durch Einstellen eines Verhältnisses der Sprachkomponente zur Nicht-Sprachkomponente und zum Umgebungsrauschsignal zu verbessern, wobei das Verhältnis durch Anlegen einer Verstärkungssteigerung an die Sprachkomponente und/oder Anlegen der Dämpfungssteigerung an die Nicht-Sprachkomponente eingestellt wird.


 
6. System nach Anspruch 5, wobei die Verhältnis-Einstelleinheit weiter konfiguriert ist, um das Verhältnis der Sprachkomponente zur Nicht-Sprachkomponente und das Umgebungsrauschsignal während eines Sprachabschnitts einzustellen, wobei der Sprachabschnitt mindestens einen Teil der Sprachkomponente enthält.
 
7. System nach einem der Ansprüche 5 bis 6, wobei die Berechnungseinheit für die erste Metrik weiter konfiguriert ist, um die erste Metrik für ein Frequenzband des Audiosignals zu berechnen, und
wobei die Erhaltungseinheit für die zweite Metrik weiter konfiguriert ist, um die zweite Metrik zumindest teilweise basierend auf dem Frequenzband zu erhalten.
 
8. System nach einem der Ansprüche 5 bis 7, wobei die die Verständlichkeit verbessernde Einheit umfasst:
eine Steigerungs-Einschränkungseinheit, die konfiguriert ist, um die bestimmte Verstärkungssteigerung und/oder die Dämpfungssteigerung basierend auf der Lautstärke des Umgebungsrauschsignals zu begrenzen, bevor die Verstärkungssteigerung auf die Sprachkomponente angewendet wird und/oder die Dämpfungssteigerung auf die Nicht-Sprachkomponente angewendet wird.
 
9. Computerprogrammprodukt zur Verbesserung der Verständlichkeit von Sprachinhalten in einem Audiosignal, wobei das Computerprogrammprodukt konkret auf einem nichtflüchtigen, computerlesbaren Medium gespeichert ist und maschinenausführbare Anweisungen umfasst, die, wenn sie ausgeführt werden, die Maschine veranlassen, das Verfahren nach einem der Ansprüche 1 bis 4 durchzuführen.
 


Revendications

1. Procédé pour améliorer l'intelligibilité d'un contenu vocal dans un signal audio, le signal audio contenant une composante vocale et une composante non vocale, la composante vocale contenant le contenu vocal, le procédé comprenant :

l'obtention d'une force sonore de référence du signal audio, dans lequel la force sonore de référence du signal audio est la force sonore du signal audio sans un signal de bruit environnemental ;

l'ajustement d'une force sonore partielle du signal audio à la force sonore de référence ;

le calcul d'une première mesure métrique sous la forme d'un rapport signal sur bruit de la composante vocale sur la composante non vocale du signal audio ajusté ;

l'obtention d'une seconde mesure métrique qui est un rapport signal sur bruit de référence de la composante vocale sur la composante non vocale et le signal de bruit environnemental tel que dérivé d'une norme d'intelligibilité vocale ;

le fait de dériver un gain d'amplification pour amplifier la composante vocale et/ou un gain d'atténuation pour atténuer la composante non vocale sur la base de la plus petite parmi la première et la seconde mesure métrique ; et

l'amélioration de l'intelligibilité du contenu vocal par l'ajustement d'un rapport de la composante vocale sur la composante non vocale et le signal de bruit environnemental, dans lequel le rapport est ajusté par l'application du gain d'amplification à la composante vocale et/ou l'application du gain d'atténuation à la composante non vocale.


 
2. Procédé selon la revendication 1, dans lequel le rapport de la composante vocale sur la composante non vocale et le signal de bruit environnemental est ajusté pendant une section vocale, la section vocale contenant au moins une partie de la composante vocale.
 
3. Procédé selon la revendication 2, dans lequel la première mesure métrique est calculée pour une bande de fréquences du signal audio, et
dans lequel la seconde mesure métrique est obtenue au moins partiellement sur la base de la bande de fréquences.
 
4. Procédé selon l'une quelconque des revendications 1 à 3, comprenant en outre :
le fait de limiter le gain d'amplification et/ou gain d'atténuation déterminés sur la base de la force sonore du signal de bruit environnemental avant l'application du gain d'amplification à la composante vocale et/ou l'application du gain d'atténuation à la composante non vocale.
 
5. Système pour améliorer l'intelligibilité d'un contenu vocal dans un signal audio, le signal audio contenant une composante vocale et une composante non vocale, la composante vocale contenant le contenu vocal, le système comprenant :

une unité d'obtention de force sonore de référence configurée pour obtenir une force sonore de référence du signal audio dans lequel la force sonore de référence du signal audio est la force sonore du signal audio sans un signal de bruit environnemental ;

une unité d'ajustement de force sonore configurée pour ajuster une force sonore partielle du signal audio à la force sonore de référence du signal audio ;

une unité de calcul de première mesure métrique configurée pour calculer une première mesure métrique sous la forme d'un rapport signal sur bruit de la composante vocale sur la composante non vocale du signal audio ajusté ;

une unité d'obtention de seconde mesure métrique configurée pour obtenir une seconde mesure métrique qui est un rapport signal sur bruit de référence de la composante vocale sur la composante non vocale et le signal de bruit environnemental tel que dérivé d'une norme d'intelligibilité vocale ;

une unité de détermination de gain configurée pour dériver un gain d'amplification pour amplifier la composante vocale et/ou un gain d'atténuation pour atténuer la composante non vocale sur la base de la plus petite parmi la première et la seconde mesure métrique ; et

une unité d'amélioration d'intelligibilité configurée pour améliorer l'intelligibilité de la composante vocale par l'ajustement d'un rapport de la composante vocale sur la composante non vocale et le signal de bruit environnemental, dans lequel le rapport est ajusté par l'application du gain d'amplification à la composante vocale et/ou l'application du gain d'atténuation à la composante non vocale.


 
6. Système selon la revendication 5, dans lequel l'unité d'ajustement de rapport est configurée en outre pour ajuster le rapport de la composante vocale sur la composante non vocale et le signal de bruit environnemental pendant une section vocale, la section vocale contenant au moins une partie de la composante vocale.
 
7. Système selon l'une quelconque des revendications 5 et 6, dans lequel l'unité de calcul de première mesure métrique est configurée en outre pour calculer la première mesure métrique pour une bande de fréquences du signal audio, et
dans lequel l'unité d'obtention de seconde mesure métrique est configurée en outre pour obtenir la seconde mesure métrique au moins partiellement sur la base de la bande de fréquences.
 
8. Système selon l'une quelconque des revendications 5 à 7, dans lequel l'unité d'amélioration d'intelligibilité comprend :
une unité de limitation de gain configurée pour limiter le gain d'amplification et/ou gain d'atténuation déterminés sur la base de la force sonore du signal de bruit environnemental avant l'application du gain d'amplification à la composante vocale et/ou l'application du gain d'atténuation à la composante non vocale.
 
9. Produit de programme informatique pour améliorer l'intelligibilité d'un contenu vocal dans un signal audio, le produit de programme informatique étant stocké de manière tangible sur un support non transitoire lisible par ordinateur et comprenant des instructions exécutables par une machine qui, quand elles sont exécutées, amènent la machine à mettre en oeuvre le procédé selon l'une quelconque des revendications 1 à 4.
 




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