A METHOD OF MODIFYING AUDIO SIGNAL FREQUENCY AND SYSTEM FOR MODIFYING AUDIO SIGNAL FREQUENCY

Abstract

 A method of modifying the sound signal frequency is characterized by that after acquiring a sample audio signal (SP) from the electronic device (UE) two, preferably three copies are formed wherein the first copy of the audio signal sample (S1) is analyzed by genre, a second copy of the audio signal sample (S2) is subjected to modification towards the optimal sound, and the third copy of the audio signal sample (S3) is treated as a reference signal. These copies are subjected to the process of modification which take into account the frequency characteristic of the room.
A system for modifying the sound signal frequency consisting of the sound buffer (BD), which is connected via a genre classification unit (UG) to control unit (US), which is connected to modification unit (UM) is characterized by that the audio buffer (BD) is connected to modification unit (UM), and to control unit (US) the room phonic analysis unit (UA) is attached. The control unit (US) is connected to modification unit (UM) via first frequency modifier (7) compensating parameters of the room acoustics and via the second frequency modifier (8) compensating genre. The first frequency modifier (7) is connected to the second frequency modifier (8) which is connected via the percentile analyzer (9) to the signal amplifier (10). A signal amplifier is connected to the block export (BE), the sound buffer (BD) is also connected directly to the first frequency modifier (7), the second frequency modifier (8) and to the signal amplifier (10) which is connected to the export block (BE).

Description

[0001] The present invention relates to a method for modifying the frequency of audio signal and system for modifying the frequency of audio signal, intended in particular for use in listening rooms and recording studios.

[0002] US Patent 8531602 discloses arrangement whereby the system detects playing a song, and then replaces it with the same as to the content but higher quality. According to this solution, the system searches the reference base and when it finds a suitable sound, it mixes the original channel with a higher resolution file. At the mixing step a suitable time delay is selected and the phase of smart signals matching occurs. The entire procedure is performed automatically and the user does not need to interfere in the process of improving sound quality. The disadvantage of this solution is that the solution does not apply to mobile devices, where low-quality transducers and loudspeakers are used.

[0003] An embodiment for improving the sound quality in the mobile devices is described in U.S. Patent 20130163784. In this embodiment, a method of improving the quality of playback of low frequencies is based on a single default gain setting, which significantly reduces the effectiveness of the solution, because each track is applied the same sound enhancement parameters

[0004] Known from the European patent EP 2384025, the embodiment relates to an intelligent sound amplification system. The system listens to the playback contents using an external device, and then adjusts the parameters to improve the sound to the collected content. The device receives audio signals and produces improved audio signals. Amplifying the sound signal is based on sound samples from the speaker of the mobile device. The main limitation of this system is the ability to improve the sound of mobile devices only when playing sound from the built-in speakers and in good acoustic conditions. When playing through headphone, the said embodiment is idle. If sounds are played in poor acoustic conditions, there is distortion in the intercepted signal, resulting in inaccurate signal modification and consequently introducing additional distortion to the signal.

[0005] US Pat 20130178963 discloses a method for recognizing musical instruments in audio recording and improving their sound. The main application of the embodiment is to use it in large collections of digital works.

[0006] The Polish patent application P.408563 presents and embodiment to modify the audio signal frequency, which is characterized in that a sound sample is acquired from the electronic device, from which the three copies are created, the first copy of the audio signal sample is analyzed for genre, the second copy of the audio signal sample is subjected to modifications towards the best sound, while the third copy of the audio signal sample is treated as a reference signal.

[0007] A method of modifying the sound signal frequency, based on acquisition of sample sound signal from the device, from which two, preferably three copies are created is characterized according to the invention in that the control system implements a first algorithm that analyzes the frequency characteristic of the room, which is obtained in a phonic analysis and analyzing genre information acquired from the genre classification unit. The first copy of sound signal sample, after identifying the genre in a genre classification unit, is sent to the control unit, in which, depending on the frequency characteristics of the room, the parameters are adjusted to modify frequency of audio signal and a first control signal is generated containing information about frequency response of the room, and a second control signal containing information about the type of music. The second copy of audio signal sample is modified in modification unit in which with using the first frequency modifier, using the first control signal, and using the implemented second algorithm for realizing method to control frequency multiband filter customized in accordance with a control signal 1, reference is made to the reference signal, and the acquired first modified audio signal is directed to the second frequency modifier. In the second frequency modifier, using the second control signal, the signal is tuned to the genre using a third algorithm implementing the method of controlling the multiband filter customized in accordance with a control signal 2, and reference is made to the reference signal. The second modified sound signal is fed to the percentile analyzer, which measures the statistical characteristic. The resulting signal of the statistical information together with a second modified sound signal is fed to a signal amplifier, wherein on the basis of statistical information SST the volume is leveled, and the resulting output signal is fed to exporting block.

[0008] In an embodiment of the invention, in a first frequency modifier using the second algorithm, it obtains a first reference signal containing information about the level of modification of the first modified audio signal and the second frequency modifier using the third algorithm, it obtains a second reference signal containing information about the level of modification of the second modified signal sound. From the amplifier a third reference signal is acquired containing information about the sound volume. Then, the first reference signal, the second reference signal and a third reference signal are fed to modification control unit. In the modification control unit, wherein using a fourth implemented algorithm for performing the control of level of signal modification, it analyzes the signal quality of the result. In case of insufficient quality, as a differential signal is fed to the control unit.

[0009] In another embodiment, a second sample of the audio signal is directed first to the modification decision unit, wherein in the case of large content of audio content, the signal is fed to the tracks separator. In the separator, the audio signal is divided into n-tracks. Each of the n-tracks is subjected to a separate modification in a modification unit, and a separate modification control unit. N-modified tracks are merged in the integrator and the resulting signal is fed to export block.

[0010] A system for modifying audio signal frequency comprising a sound buffer, which is connected via genre classification unit with a control unit which is connected to a modification unit is characterized in that the audio buffer is connected to modification unit, and control unit is coupled with room phonic analysis unit. The control unit is connected to modification unit via first frequency modifier compensating frequency characteristics acoustics of the room and via the second frequency modifier compensating the genre. The first frequency modifier is connected to the second frequency modifier which is connected via the percentile analyzer to the signal amplifier. Signal amplifier is connected to the export block. At the same time the sound buffer is also connected directly to the first frequency modifier, the second frequency modifier and with a signal amplifier which is connected to the export block.

[0011] In another embodiment of invention, modification unit is connected to a modification control unit via first frequency modifier, second frequency modifier and signal amplifier, while the modification control unit is connected to control unit.

[0012] In another embodiment of invention, decision unit is located on the input for the modification unit, which is connected to track separator, and the track integrator is located on the output of the modification unit.

[0013] The invention is based on an intelligent use of the control sound utilizing the musical genres and improve the sound quality, depending on the device used.

[0014] Modification of the audio signal, according to the invention is carried out according to the genre and taking into account the acoustics of the room in which the sound is played. The modification signal may take place in the entire audio signal or the individual instrumental tracks. When carrying out modifications to the signal frequency, the system recognizes music genre from the processed audio signal, also analyzes the acoustic environment, adapting the process to the current acoustic environment in the room.

[0015] The present invention, by utilizing the phenomena of perceptual frequency amplification, restores bands, which physically could not be restored. The invention extends the possibility to improve the sound quality of intelligent diversification of gain parameters depending on the genre, making it possible to achieve less distortion in the audio processing.

[0016] The invention is presented using embodiments where,:

fig. 1 presents a general block diagram of the system,

fig. 2 presents a basic method of audio signal frequency modification,

fig. 3 presents a more detailed method for modification,

fig. 4 presents a more detailed method of modification of individual sound tracks.

Embodiment 1

[0017] As shown in fig. 2 sound source UE is connected to the audio buffer 1.

[0018] A sound signal frequency modification unit is composed of the sound buffer BD, which is connected via genre classification unit UG with the control unit US, which is connected to modification unit UM.

[0019] Genre classification unit UG consists of a filter block 1 that filters the signal in a manner appropriate to the applied parameters, parameterization block 2, the signal data redundancy reducer 3, genres classifier 4. The result of the recognition genre is transmitted to the control unit US.

[0020] Room phonic analysis unit UA is connected to control unit US is included in audio analysis system for UA rooms Room phonic analysis unit UA is comprised of the impulse response recorder 5 and Fourier transform analyzer 6.

[0021] Control unit US is connected with modification unit UM via the first frequency modifier 7 compensating acoustic characteristics of the room and through the second frequency modifier 8 compensating genre.

[0022] The first frequency modifier 7 is connected to a second frequency modifier 8 which is connected to the percentile analyzer 9 with a signal amplifier 10 which is connected to the export block BE. The first frequency modifier 7, second frequency modifier 8 and amplifier 10 are also connected directly to the sound buffer BD.

[0023] Control unit US implements the first algorithm ALI analyzing frequency characteristics of the room and analyzing information acquired from the genre classification. The first modifier implements a second algorithm AL2 implementing the method of controlling a multi-band frequency filter, customized in accordance with control signal 1. The second modifier implements the third algorithm AL3 realizing the filter control frequency multiband customized in accordance with a control signal 2.

[0024] A sample audio signal SP is acquired from the electronic device EU, which is used to create two copies. The first copy of the audio signal sample S1 is analyzed in terms of genre, a second copy of the audio signal sample S2 is subjected to modifications towards optimal sound.

[0025] With the help of room phonic analysis unit, frequency response of the room is measured, and characteristics signal SCH is generated. Characteristics signal SCH is acquired from impulse response S01 which is acquired from impulse response recorder 5 and processed by Fourier transform analyzer 6.

[0026] The first copy of the audio signal sample SI, after identifying the genre in genre classification unit UG, is sent to the control unit US, where depending on the frequency characteristics of the room, acquired from in the form of signal characteristics SCH and an information signal S1I, frequency modification parameters of audio signal S2 are adjusted. Control unit US generates a first control signal SS1 containing information about frequency response of the room, and the second control signal SS2, containing information about the genre of music.

[0027] The second copy of audio signal sample S2 is modified in a modification unit UM, in which, with the first frequency modifier 7, using the first control signal SS1 and through implementing a second algorithm AL2, reference is made to the reference signal S3. The first resulting modified audio signal S2Z1 is directed to second frequency modifier 8. The second frequency modifier 8, using the second control signal SS2, tunes the signal to the genre using a third algorithm AL3 and reference is made to the reference signal S3. The second modified audio signal, S2Z2 is directed to the percentile analyzer 9, which measures statistical characteristic. The resulting signal SST of statistical information together with a second modified sound signal S2Z2 are fed to signal amplifier 10, which in comparison to the reference signal S3 the intensity of the second modified sound signal S2Z2 is measured. Based on this measurement volume is leveled, and the resulting output signal SK is fed to the exporting block BE. Embodiment 2

[0028] Three copies of the signal SP are acquired from the data buffer. As shown in fig. 3 unit is constructed as described in Embodiment 1, except that the modification unit UM is connected to a modification control unit UK through the first frequency modifier 7, the second frequency modifier 8 and a signal amplifier 10, while modification control unit UK is connected to control unit US

[0029] Modification control unit UK implements the fourth algorithm AL4 that controls the level of the signal modification by analyzing the output signal quality.

[0030] First frequency modifier 7, using a second algorithm AL2, acquires first reference signal S01 containing information about the level of the first modification of the modified audio signal S2Z1. Second frequency modifier 8, using the third algorithm AL3, acquires second reference signal S02 containing the information about the level of the second modification of the modified audio signal S2Z2. The amplifier 10 provides the third reference signal S03 that contains information about sound level.

[0031] First reference signal S01, second reference signal S02 and third reference signal S03 are directed to modification control unit UK, wherein implemented uses fourth algorithm AL4 to analyze the quality of the output signal SK. In case of insufficient quality, as a differential signal SR is fed to the control system the US. Then, after achieving the right quality in the control unit US, this signal as the result signal SK is directed to exporting block BE.

Embodiment 3

[0032] As shown in fig. 4 the system is constructed as described in Embodiment 2, except that the input to the modification unit UM is located in the decision unit UD which is connected to the paths separator 11, and on the output of the modification unit UM the track integrator 12 is located.

[0033] A second sample of the audio signal S2 is directed in the first instance to the decision unit UD, in the case of wide audio content consisting of three instrumental tracks, this signal is fed to the paths separator 11, where it is divided into three tracks S2P. Each of the separate tracks are subjected to separate modifications on the modification unit UM and a separate control modifications on modification control unit UK. Modified tracks S2Z are merged in integrator 12 and fed to export block BE.

Claims

1. A method of modifying the sound signal frequency, which is characterized in that a sample audio signal (SP) is acquired from the electronic device (UE), from which two, preferably three copies are formed wherein the first copy of the audio signal sample (S1) is analyzed by genre, a second copy of the audio signal sample (S2) is subjected to modification towards the optimal sound, and the third copy of the audio signal sample (S3) is treated as a reference signal, characterized in that the control unit (US) implements the first algorithm ( ALI) analyzing the frequency characteristic of the room, which is acquired in the phonic analysis unit (AU) and analyzing information on a genre obtained from the classification of genres (UG) and the first copy of the audio signal sample (S1), after identifying the genre in genre classification (UG) is sent to the control unit (US), in which, depending on the frequency space, adapts the parameters to modify the frequency acoustic signal (S1) and generates first control signal (SS1) containing information about the frequency response of the room, and a second control signal (SS2), comprising information on the genre of music, while a second copy of the audio signal sample (S2) are modified in a modification unit (UM), wherein by first frequency modifier (7) using first control signal (SS1) and using the implemented second algorithm (AL2) for realizing the control of frequency multiband filter customized in accordance with a control signal 1, and reference is made to the reference signal (S3), and the resulting first modified audio signal (S2Z1) is sent to second frequency modifier (8), wherein the second frequency modifier (8), using the second control signal (SS2), tunes the signal to the genre using the implemented third algorithm (AL3) for realizing the control of frequency multiband filter customized in accordance with a control signal 2 and reference is made to the reference signal (S3) and a second modified sound signal (S2Z2) is fed to the percentile analyzer (9), which measures the statistical characteristic, then the signal containing statistical information (SST) and a second modified sound signal (S2Z2) are fed to signal amplifier (10), wherein in comparison to the reference signal (S3), the intensity of the second modified sound signal (S2Z2), and based on this measurement leveled volume, and the resulting output signal (SK) is fed to export block (BE).

2. A method according to claim. 1 characterized in that the first frequency modifier (7) using the second algorithm (AL2) first reference signal (S01) is acquired containing information about the level of modification of the first modified audio signal (S2Z1), and a second frequency modifier (8) using a third algorithm (AL3) the second reference signal (S02) is acquired containing information about the level of modification of the second modified sound signal (S2Z2) and the amplifier (10) provides a third reference signal (S03) containing information on the sound level, after which the first reference signal (S01), second reference signal (S01) and third reference signal (S03) are directed to the modification control unit (UK), wherein by means of fourth algorithm (AL4), it realizes controlling of the level of the signal modification by analyzing the quality of the resulting signal (SK), and in case of insufficient quality as differential signal (SR) is fed to the control unit (US).

3. A method according to claim 2, characterized in that the second audio signal sample (S2) is directed first to the modification decision unit (UD), wherein in case of wide content of the audio signal it is fed to the tracks separator (11) where tracks are divided into n-tracks (S2P), wherein each of the n-tracks is subjected to separate modification in modification unit (UM) and separate control modifications in modification control unit (UK) and n-modified tracks (S2Z) are merged in integrator (12) and the output signal (SK) is directed to an exporting unit (BE).

4. A system for modifying the sound signal frequency consisting of the sound buffer (BD), which is connected via a genre classification unit (UG) to control unit (US), which is connected to modification unit (UM), characterized in that the audio buffer (BD) is connected to modification unit (UM), and to control unit (US) is attached room phonic analysis unit (UA), wherein the control unit (US) is connected to modification unit (UM) via first frequency modifier (7) compensating parameters of the room acoustics and via the second frequency modifier (8) compensating genre, the first frequency modifier (7) is connected to second frequency modifier (8) which is connected via the percentile analyzer (9) with a signal amplifier (10) which is connected to the block export (BE), the sound buffer (BD) is also connected directly to the first frequency modifier (7), the second frequency modifier (8) and with signal amplifier (10) which is connected export block (BE).

5. The system according to claim. 4, characterized in that the modification unit (UM) is connected to the modification control unit (UK) via first frequency modifier (7), second frequency modifier (8) and amplifier (10) and the modification control unit (UK) is connected with a control unit (US).

6. The system according to claim. 5, characterized in that at the input to the modification unit (UM) there is decision unit (UD), which is connected to the track separator (11), and on the output of the modification unit (UM) there is track integrator (12).

Drawing