CN111179958A - Method and system for inhibiting late reverberation of voice - Google Patents
Method and system for inhibiting late reverberation of voice Download PDFInfo
- Publication number
- CN111179958A CN111179958A CN202010016846.XA CN202010016846A CN111179958A CN 111179958 A CN111179958 A CN 111179958A CN 202010016846 A CN202010016846 A CN 202010016846A CN 111179958 A CN111179958 A CN 111179958A
- Authority
- CN
- China
- Prior art keywords
- reverberation
- voice
- spectrum characteristic
- frequency spectrum
- speech
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 230000002401 inhibitory effect Effects 0.000 title claims abstract description 9
- 238000001228 spectrum Methods 0.000 claims abstract description 56
- 230000001629 suppression Effects 0.000 claims abstract description 35
- 238000004364 calculation method Methods 0.000 claims abstract description 19
- 230000005540 biological transmission Effects 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 8
- 230000003595 spectral effect Effects 0.000 claims abstract description 8
- 238000011410 subtraction method Methods 0.000 claims abstract description 4
- 230000008569 process Effects 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 5
- 238000009432 framing Methods 0.000 claims description 3
- 230000007547 defect Effects 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 210000005069 ears Anatomy 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- G10L2021/02082—Noise filtering the noise being echo, reverberation of the speech
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Circuit For Audible Band Transducer (AREA)
Abstract
The invention discloses a method and a system for inhibiting late reverberation of voice, and belongs to the technical field of voice processing. Firstly, collecting indoor reverberation voice signals and extracting reverberation attenuation parameters; calculating the reverberation frequency spectrum characteristic a according to the reverberation attenuation parameter; then, calculating a voice frequency spectrum characteristic b after reverberation suppression according to the reverberation frequency spectrum characteristic a and a naive spectral subtraction method; and then, carrying out interpolation calculation on the voice frequency spectrum characteristic b, reducing the voice frequency spectrum characteristic b by utilizing the phase characteristic of the reverberation voice signal to obtain waveform voice, and outputting the waveform voice. The system comprises a collecting unit, a calculating unit, a transmission unit and a terminal, wherein the collecting unit is electrically connected with the calculating unit, and the calculating unit and the terminal are respectively connected with the transmission unit; the invention aims to overcome the defect that the voice late reverberation suppression method in the prior art cannot adapt to a time-varying reverberation environment.
Description
Technical Field
The present invention relates to the field of speech processing technologies, and in particular, to a method and a system for suppressing late reverberation of speech.
Background
Reverberation is a process that sound waves are diffusely reflected among barriers such as walls, ceilings, furniture and the like when the sound waves are transmitted indoors, the energy of the sound waves reflected once is weakened by the barriers, and the sound waves are reflected and absorbed for many times indoors and disappear finally after a sound source stops sounding.
T60 is used in acoustics to estimate the reverberation duration of a room, typically a small and medium conference room, which does not exceed 1 s. In the real-time conference communication equipment, a sound source emits sound, the sound source directly picked up by a microphone is called direct sound (the sound source directly points to the microphone without any reflection, and is picked up as reverberation by the microphone after being reflected), after the direct sound arrives, the reverberation arriving within 80ms is called early reverberation, and the reverberation arriving after 80ms is called late reverberation. The early reverberation can enhance the saturation of the voice, so that the human voice can be more saturated. Late reverberation can reduce the intelligibility of speech, affect the subjective experience of human ears and reduce the recognition rate of a speech recognition model.
In the prior art, the reverberation is mainly suppressed by a traditional method, and the method can be divided into the following three types according to the used technology: a reverberation suppression method based on beam forming, inverse filtering and neural network. The reverberation suppression method of beam forming is easy to implement, but is difficult to adapt to a time-varying reverberation environment; the reverberation suppression method based on the inverse filtering technology also has the advantages that accurate parameters cannot be updated in real time due to the time-varying reverberation environment; in recent years, a reverberation suppression method based on a neural network is applied, but due to the diversity of reverberation types and the limitation caused by training data, models and the like, the accuracy of the reverberation suppression method is low, and the reverberation suppression method is difficult to apply to deployment and real-time conference communication equipment.
Some related technologies are disclosed in the prior art, such as the name of the invention creation: the scheme discloses a processing method and a processing device for reverberation suppression (application date: 2019, 1, 30 and application number: 201910090031.3). The method comprises the following steps: acquiring sound data to be processed, wherein the sound data to be processed is first sound data containing reverberation; processing the first sound data according to a similarity matrix to obtain second sound data subjected to preliminary reverberation suppression, wherein the similarity matrix is obtained by pre-training; processing the second sound data according to a Wavenet network model, wherein the Wavenet network model is obtained by pre-training; and acquiring the output third sound data from the Wavenet network model. The scheme solves the problems of low accuracy and limited application scene of a reverberation suppression algorithm in the related technology. However, the disadvantages of this solution are: in the scheme, training data of 20h cannot meet diversified and rich reverberation environments, the processing dependency of reverberation data on a similarity matrix is high, and once a reverberation scene with an incompatible similarity matrix appears, abnormal conditions are easy to appear.
In summary, how to adapt the late reverberation suppression method to the time-varying reverberation environment is an urgent problem to be solved in the prior art.
Disclosure of Invention
1. Problems to be solved
The invention aims to overcome the defect that a late reverberation suppression method of voice in the prior art cannot adapt to a time-varying reverberation environment, and provides a late reverberation suppression method and a late reverberation suppression system of voice, which can adapt to the time-varying reverberation environment, stably suppress late reverberation, and further enhance the effect of late reverberation suppression.
2. Technical scheme
In order to solve the problems, the technical scheme adopted by the invention is as follows:
the invention discloses a method for inhibiting late reverberation of voice, which comprises the steps of collecting indoor reverberation voice signals and extracting reverberation attenuation parameters; calculating the reverberation frequency spectrum characteristic a according to the reverberation attenuation parameter; then, calculating a voice frequency spectrum characteristic b after reverberation suppression according to the reverberation frequency spectrum characteristic a and a naive spectral subtraction method; and then, carrying out interpolation calculation on the voice frequency spectrum characteristic b, reducing the voice frequency spectrum characteristic b by utilizing the phase characteristic of the reverberation voice signal to obtain waveform voice, and outputting the waveform voice.
Further, the specific process of extracting the reverberation attenuation parameter includes: windowing and framing the reverberation voice signal and performing short-time Fourier transform; then calculating the amplitude spectrum of each frame and uniformly dividing the amplitude spectrum into H frequency bands; then, calculating the mean value and the maximum value of each frequency band to obtain H-dimensional sub-frequency band characteristics and H maximum values; and then, normalizing the H-dimensional sub-band characteristics by using the H maximum values to obtain H reverberation attenuation parameters.
Further, the reverberation spectrum characteristic a is calculated by the following formula:
Wherein
The estimated reverberation frequency spectrum characteristic a of the current frame is represented, X represents the frequency spectrum characteristic of the input signal, i represents the frame index of the current reverberation frequency spectrum characteristic, j represents the frame index of the reverberation attenuation parameter, c represents the time from the sound source to the microphone, and d represents the length of the reverberation attenuation parameter.
Further, the speech spectral feature b is interpolated by using a linear interpolation method.
Further, the normalization process is performed by the following formula:
D＝Xsubband/M
wherein D represents the normalized attenuation parameter, XsubbandThe H-dimensional subband characteristic is represented, and M represents the maximum value corresponding to the H-dimensional subband.
Further, the speech spectrum feature b after reverberation suppression is calculated by using the following formula:
wherein
Representing the speech spectral feature b after reverberation suppression.
Further, interpolating the H-dimensional subband characteristic of the speech spectrum characteristic b to obtain a plurality of frequency points, and interpolating the speech spectrum characteristic b by using the following formula:
wherein f represents the number of the frequency point, begin represents the first frequency point of the current sub-band, end represents the last frequency point of the current sub-band, and y represents the value of each frequency point.
Further, the phase characteristics of the reverberation voice signal are used for carrying out inverse Fourier transform to restore the voice spectrum characteristics b to obtain waveform voice.
The invention relates to a voice late reverberation suppression system which comprises a collection unit, a calculation unit, a transmission unit and a terminal, wherein the collection unit is electrically connected with the calculation unit, and the calculation unit and the terminal are respectively connected with the transmission unit, wherein the calculation unit comprises a memory and a processor, the memory is connected with the processor, a program is stored in the memory, the program is used for realizing the voice late reverberation suppression method, and the processor is used for executing the program to output waveform voice.
Furthermore, the acquisition unit comprises a sound collector and a signal converter, the sound collector is electrically connected with the signal converter, and the signal converter is electrically connected with the calculation unit.
3. Advantageous effects
Compared with the prior art, the invention has the beneficial effects that:
according to the method for inhibiting the late reverberation of the voice, disclosed by the invention, the late reverberation of the voice can be inhibited by combining voice attenuation parameters, and the method can adapt to a time-varying reverberation environment, so that the effect of inhibiting the late reverberation can be enhanced; in addition, the extraction of the voice attenuation parameters is originated from the voice data with reverberation, and corresponding parameter estimation can be carried out according to the change of the reverberation voice data, so that the conditions of over-reduction or insufficient inhibition are effectively avoided; the method is easy to realize and high in robustness, and can further improve the recognition rate of the voice. The late reverberation suppression system for the voice can suppress late reverberation of the voice, can adapt to a time-varying reverberation environment, and further can provide a good voice environment for a real-time conference.
Drawings
FIG. 1 is a flow chart of a method for suppressing late reverberation of speech according to the present invention;
fig. 2 is a schematic diagram of the structure of the late reverberation suppression system of the present invention.
The reference numerals in the schematic drawings illustrate: 100. a collection unit; 200. a calculation unit; 300. a transmission unit; 400. and (4) a terminal.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some embodiments of the present invention, but not all embodiments; moreover, the embodiments are not relatively independent, and can be combined with each other according to needs, so that a better effect is achieved. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
For a further understanding of the invention, reference should be made to the following detailed description taken in conjunction with the accompanying drawings and examples.
Example 1
Referring to fig. 1, a method for suppressing late reverberation of speech according to the present invention includes the following steps:
collecting indoor reverberation voice signals and extracting reverberation attenuation parameters; the specific process of extracting the reverberation attenuation parameter is as follows:
(1) windowing and framing the reverberation voice signal and performing short-time Fourier transform;
(2) calculating the magnitude spectrum of each frame and uniformly dividing the magnitude spectrum into H frequency bands; in this embodiment, the magnitude spectrum is uniformly divided into 40 frequency bands, and it should be noted that the process of calculating the magnitude spectrum of each frame is as follows: windowing a frame, performing short-time Fourier transform to obtain a plurality of complex values, and then taking the sum of squares of the complex values to obtain a magnitude spectrum of the frame; in this embodiment, 20ms is taken as a frame, the step length is 10ms, short-time fourier transform is performed to obtain 160 complex values, and then the square value is taken to obtain the amplitude spectrum of the frame;
(3) calculating the mean value and the maximum value of each frequency band to obtain H-dimensional sub-frequency band characteristics and H maximum values; the embodiment obtains 40-dimensional subband characteristics and 40 maximum values; calculating the mean value of each frequency band by using the following formula to obtain the H-dimensional sub-band characteristics:
calculating the maximum value of each frequency band by using the following formula to obtain H maximum values;
wherein i and k represent subband numbers within a band range, and x represents a subband value; for example, the range of the first frequency band is [1,4], the first dimension subband characteristic is obtained by adding 4 subband values and averaging, and the maximum value of the 4 subband values is taken as the maximum value of the first frequency band.
(4) And carrying out normalization processing on the H-dimensional sub-band characteristics by utilizing the H maximum values to obtain H reverberation attenuation parameters. It is worth mentioning that the normalization process is performed by the following formula:
D＝Xsubband/M
wherein D represents the normalized attenuation parameter, XsubbandThe H-dimensional subband characteristic is represented, and M represents the maximum value corresponding to the H-dimensional subband. Example XsubbandRepresenting a 40-dimensional subband characteristic, and M represents a maximum corresponding to the 40-dimensional subband. It should be noted that 1 frame represents 10ms, d frame attenuation parameters need to be acquired at one time in the present invention, d represents the length of the reverberation attenuation parameter, and d is equal to 100 in this embodiment.
The reverberation attenuation parameters can be extracted through the steps, and the reverberation voice signals are processed through the reverberation attenuation parameters, so that the voice attenuation parameters can be calculated and updated in real time, and the time-varying reverberation environment can be better adapted; further suppression of late reverberation may be achieved.
Calculating reverberation frequency spectrum characteristics a according to the reverberation attenuation parameters; specifically, the H-dimensional subband characteristic in the above step is adopted, and the reverberation spectrum characteristic a is calculated by the following formula:
Wherein
Representing the estimated reverberation frequency spectrum characteristic a of the current frame, X representing the frequency spectrum characteristic of an input signal, i representing the frame index of the current reverberation frequency spectrum characteristic, j representing the frame index of a reverberation attenuation parameter, c representing the time from the sound source to the microphone, and d representing the length of the reverberation attenuation parameter; in this example, c is 1.
Calculating a voice frequency spectrum characteristic b after reverberation suppression according to the reverberation frequency spectrum characteristic a and a naive spectral subtraction method; calculating the voice frequency spectrum characteristic b after the reverberation suppression by using the following formula:
wherein
Representing the speech spectral feature b after reverberation suppression.
And performing interpolation calculation on the voice frequency spectrum characteristic b, specifically, performing interpolation on the H-dimensional subband characteristic of the voice frequency spectrum characteristic b to obtain a plurality of frequency points, and performing interpolation on the voice frequency spectrum characteristic b by using a linear interpolation method. In this embodiment, the interpolation is performed on the 40-dimensional subband characteristic of the speech spectrum characteristic b to obtain 161 frequency points, and a specific interpolation formula is as follows:
wherein f represents the frequency point number, begin represents the first frequency point of the current sub-band, end represents the last frequency point of the current sub-band, and y represents the value of each frequency point.
It should be noted that in this embodiment, the 40-dimensional subbands are interpolated into 161 frequency bins, each dimension subband represents a range of 4 frequency bins, that is, the 1-dimensional subband represents 1 to 4 frequency bins, the 2-dimensional subband represents 4 to 8 frequency bins, and so on, and the last one-dimensional subband represents 156 to 160. Except for the 1 st dimensional sub-band, the last frequency point of the front one-dimensional sub-band is used as the starting frequency point of the rear one-dimensional sub-band. Wherein, the value of 1 frequency point is 0, the value of 4 frequency points is the value of 1 frequency sub-band, the value of 8 frequency points is the value of 2 frequency sub-band, and so on. Taking 10ms as a frame to perform Fourier transform in an audio signal with the sampling rate of 16000Hz, and further obtaining fixed 161 frequency points.
And restoring the voice frequency spectrum characteristic b by using the phase characteristic of the reverberation voice signal to obtain waveform voice, and outputting the waveform voice. Specifically, the phase characteristic of the reverberation voice signal is used for carrying out inverse Fourier transform to restore the voice spectrum characteristic b to obtain waveform voice.
According to the method for inhibiting the late reverberation of the voice, disclosed by the invention, the late reverberation of the voice can be inhibited by combining voice attenuation parameters, and the method can adapt to a time-varying reverberation environment, so that the effect of inhibiting the late reverberation can be enhanced; in addition, it is worth explaining that the extraction of the voice attenuation parameters is originated from the voice data with reverberation, and corresponding parameter estimation can be performed according to the change of the reverberation voice data, so that the conditions of over-reduction or insufficient suppression are effectively avoided; the method is easy to realize and high in robustness, and can further improve the recognition rate of the voice.
Referring to fig. 2, the late reverberation suppression system of the present invention includes an acquisition unit 100, a calculation unit 200, a transmission unit 300, and a terminal 400, wherein the acquisition unit 100 is electrically connected to the calculation unit 200, and the calculation unit 200 and the terminal 400 are respectively connected to the transmission unit 300; specifically, the collection unit 100 includes a sound collector electrically connected to a signal converter electrically connected to the calculation unit 200, and a signal converter. The sound collector is used for collecting voice, the signal converter is used for converting acoustic signals in the environment into digital signals, and the sound collector microphone in the embodiment is provided with the signal converter which is an ADC (analog-to-digital converter) hardware chip; the computing unit 200 includes a memory and a processor, the memory is electrically connected to the processor, the memory stores a program, the program is used for implementing the above-mentioned method for suppressing late reverberation of speech, the processor is used for executing the program to output waveform speech, in this embodiment, the computing unit 200 is a single chip or a computer; the transmission unit 300 is used for transmitting the data calculated by the calculation unit 200, and the transmission unit 300 is a network system for transmitting data in the embodiment; the terminal 400 is configured to play the processed audio data, and the terminal 400 is a real-time conference communication system accessing to a network. The late reverberation suppression system for the voice can suppress late reverberation of the voice, can adapt to a time-varying reverberation environment, and further can provide a good voice environment for a real-time conference.
The invention has been described in detail hereinabove with reference to specific exemplary embodiments thereof. It will, however, be understood that various modifications and changes may be made without departing from the scope of the invention as defined in the appended claims. The detailed description and drawings are to be regarded as illustrative rather than restrictive, and any such modifications and variations are intended to be included within the scope of the present invention as described herein. Furthermore, the background is intended to be illustrative of the state of the art as developed and the meaning of the present technology and is not intended to limit the scope of the invention or the application and field of application of the invention.
Claims (10)
1. A method for suppressing late reverberation of speech is characterized by comprising the following steps: comprises that
Collecting indoor reverberation voice signals and extracting reverberation attenuation parameters;
calculating reverberation frequency spectrum characteristics a according to the reverberation attenuation parameters;
calculating a voice frequency spectrum characteristic b after reverberation suppression according to the reverberation frequency spectrum characteristic a and a naive spectral subtraction method;
and carrying out interpolation calculation on the voice frequency spectrum characteristic b, reducing the voice frequency spectrum characteristic b by utilizing the phase characteristic of the reverberation voice signal to obtain waveform voice, and outputting the waveform voice.
2. The method of claim 1, wherein the step of suppressing the late reverberation of the speech comprises: the specific process for extracting the reverberation attenuation parameter comprises the following steps:
windowing and framing the reverberation voice signal and performing short-time Fourier transform;
calculating the magnitude spectrum of each frame and uniformly dividing the magnitude spectrum into H frequency bands;
calculating the mean value and the maximum value of each frequency band to obtain H-dimensional sub-frequency band characteristics and H maximum values;
and carrying out normalization processing on the H-dimensional sub-band characteristics by utilizing the H maximum values to obtain H reverberation attenuation parameters.
3. The method of claim 1, wherein the step of suppressing the late reverberation of the speech comprises: calculating the reverberation spectrum characteristic a by the following formula:
Wherein
The estimated reverberation frequency spectrum characteristic a of the current frame is represented, X represents the frequency spectrum characteristic of the input signal, i represents the frame index of the current reverberation frequency spectrum characteristic, j represents the frame index of the reverberation attenuation parameter, c represents the time from the sound source to the microphone, and d represents the length of the reverberation attenuation parameter.
4. The method of claim 1, wherein the step of suppressing the late reverberation of the speech comprises: and (4) interpolating the voice frequency spectrum characteristic b by using a linear interpolation method.
5. The method of late reverberation suppression of speech as in claim 2, wherein: the normalization process is performed by the following formula:
D＝Xsubband/M
wherein D represents the normalized attenuation parameter, XsubbandThe H-dimensional subband characteristic is represented, and M represents the maximum value corresponding to the H-dimensional subband.
6. The method of claim 3, wherein the step of suppressing the late reverberation of the speech comprises: calculating the voice frequency spectrum characteristic b after the reverberation suppression by using the following formula:
wherein
Representing the speech spectral feature b after reverberation suppression.
7. The method of claim 4, wherein the step of suppressing the late reverberation of the speech comprises: interpolating the H-dimensional sub-band characteristic of the speech audio spectrum characteristic b to obtain a plurality of frequency points, and interpolating the speech audio spectrum characteristic b by using the following formula:
wherein f represents the number of the frequency point, begin represents the first frequency point of the current sub-band, end represents the last frequency point of the current sub-band, and y represents the value of each frequency point.
8. The method for suppressing late reverberation of speech according to any of claims 1-7, wherein: and performing inverse Fourier transform on the phase characteristics of the reverberation voice signal to restore the voice spectrum characteristics b to obtain waveform voice.
9. A speech late reverberation suppression system, characterized by: the method comprises an acquisition unit, a calculation unit, a transmission unit and a terminal, wherein the acquisition unit is electrically connected with the calculation unit, the calculation unit and the terminal are respectively connected with the transmission unit, the calculation unit comprises a memory and a processor, the memory is connected with the processor, a program is stored in the memory, the program is used for realizing the method for inhibiting the late reverberation of the voice as claimed in any one of claims 1 to 8, and the processor is used for executing the program to output waveform voice.
10. The system of claim 9, wherein the collection unit comprises a sound collector and a signal converter, the sound collector is electrically connected to the signal converter, and the signal converter is electrically connected to the computing unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010016846.XA CN111179958A (en) | 2020-01-08 | 2020-01-08 | Method and system for inhibiting late reverberation of voice |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010016846.XA CN111179958A (en) | 2020-01-08 | 2020-01-08 | Method and system for inhibiting late reverberation of voice |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111179958A true CN111179958A (en) | 2020-05-19 |
Family
ID=70649356
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010016846.XA Pending CN111179958A (en) | 2020-01-08 | 2020-01-08 | Method and system for inhibiting late reverberation of voice |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111179958A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080292108A1 (en) * | 2006-08-01 | 2008-11-27 | Markus Buck | Dereverberation system for use in a signal processing apparatus |
| CN103067821A (en) * | 2012-12-12 | 2013-04-24 | 歌尔声学股份有限公司 | Method of and device for reducing voice reverberation based on double microphones |
| CN103220595A (en) * | 2012-01-23 | 2013-07-24 | 富士通株式会社 | Audio processing device and audio processing method |
| CN103262163A (en) * | 2010-10-25 | 2013-08-21 | 弗兰霍菲尔运输应用研究公司 | Echo suppression comprising modeling of late reverberation components |
| CN105580070A (en) * | 2013-07-22 | 2016-05-11 | 弗朗霍夫应用科学研究促进协会 | Apparatus and method for decoding and encoding an audio signal using adaptive spectral tile selection |
-
2020
- 2020-01-08 CN CN202010016846.XA patent/CN111179958A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080292108A1 (en) * | 2006-08-01 | 2008-11-27 | Markus Buck | Dereverberation system for use in a signal processing apparatus |
| CN103262163A (en) * | 2010-10-25 | 2013-08-21 | 弗兰霍菲尔运输应用研究公司 | Echo suppression comprising modeling of late reverberation components |
| CN103220595A (en) * | 2012-01-23 | 2013-07-24 | 富士通株式会社 | Audio processing device and audio processing method |
| CN103067821A (en) * | 2012-12-12 | 2013-04-24 | 歌尔声学股份有限公司 | Method of and device for reducing voice reverberation based on double microphones |
| CN105580070A (en) * | 2013-07-22 | 2016-05-11 | 弗朗霍夫应用科学研究促进协会 | Apparatus and method for decoding and encoding an audio signal using adaptive spectral tile selection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107479030B (en) | Frequency division and improved generalized cross-correlation based binaural time delay estimation method | |
| JP4210521B2 (en) | Noise reduction method and apparatus | |
| CN109727604A (en) | Frequency domain echo cancel method and computer storage media for speech recognition front-ends | |
| CN111833896B (en) | Voice enhancement method, system, device and storage medium for fusing feedback signals | |
| CN105869651B (en) | Binary channels Wave beam forming sound enhancement method based on noise mixing coherence | |
| CN106710601A (en) | Voice signal de-noising and pickup processing method and apparatus, and refrigerator | |
| CN108447496B (en) | Speech enhancement method and device based on microphone array | |
| CN102347028A (en) | Double-microphone speech enhancer and speech enhancement method thereof | |
| CN109727607B (en) | Time delay estimation method and device and electronic equipment | |
| CN109523999B (en) | Front-end processing method and system for improving far-field speech recognition | |
| CN110265065B (en) | Method for constructing voice endpoint detection model and voice endpoint detection system | |
| CN108010536B (en) | Echo cancellation method, device, system and storage medium | |
| CN108986832A (en) | Ears speech dereverberation method and device based on voice probability of occurrence and consistency | |
| CN105390142A (en) | Digital hearing aid voice noise elimination method | |
| CN111863015A (en) | Audio processing method and device, electronic equipment and readable storage medium | |
| CN116013344A (en) | Speech enhancement method under multiple noise environments | |
| CN106328160B (en) | Noise reduction method based on double microphones | |
| CN112201273A (en) | Noise power spectral density calculation method, system, equipment and medium | |
| CN112820312B (en) | Voice separation method and device and electronic equipment | |
| CN116612778B (en) | Echo and noise suppression method, related device and medium | |
| CN112151055A (en) | Audio processing method and device | |
| CN114023352B (en) | Voice enhancement method and device based on energy spectrum depth modulation | |
| CN111179958A (en) | Method and system for inhibiting late reverberation of voice | |
| CN111883154B (en) | Echo cancellation method and device, computer-readable storage medium, and electronic device | |
| CN113593599A (en) | Method for removing noise signal in voice signal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200519 |