CN114007157A - Intelligent noise reduction communication earphone - Google Patents
Intelligent noise reduction communication earphone Download PDFInfo
- Publication number
- CN114007157A CN114007157A CN202111260119.9A CN202111260119A CN114007157A CN 114007157 A CN114007157 A CN 114007157A CN 202111260119 A CN202111260119 A CN 202111260119A CN 114007157 A CN114007157 A CN 114007157A
- Authority
- CN
- China
- Prior art keywords
- signal
- noise
- voice
- module
- communication
- 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
- 238000004891 communication Methods 0.000 title claims abstract description 54
- 230000009467 reduction Effects 0.000 title claims abstract description 52
- 238000012545 processing Methods 0.000 claims abstract description 36
- 230000007613 environmental effect Effects 0.000 claims abstract description 27
- 238000001914 filtration Methods 0.000 claims abstract description 19
- 230000003321 amplification Effects 0.000 claims abstract description 12
- 238000003199 nucleic acid amplification method Methods 0.000 claims abstract description 12
- 230000005540 biological transmission Effects 0.000 claims abstract description 11
- 238000004458 analytical method Methods 0.000 claims abstract description 8
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000006243 chemical reaction Methods 0.000 claims abstract description 6
- 230000006870 function Effects 0.000 claims abstract description 6
- 238000010801 machine learning Methods 0.000 claims description 52
- 238000012549 training Methods 0.000 claims description 14
- 230000004044 response Effects 0.000 claims description 9
- 238000005070 sampling Methods 0.000 claims description 9
- 230000003044 adaptive effect Effects 0.000 claims description 7
- 238000009432 framing Methods 0.000 claims description 3
- 239000002184 metal Substances 0.000 claims description 3
- 238000000926 separation method Methods 0.000 claims description 3
- 238000000034 method Methods 0.000 description 16
- 238000013473 artificial intelligence Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 210000005069 ears Anatomy 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 210000000697 sensory organ Anatomy 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R1/00—Details of transducers, loudspeakers or microphones
- H04R1/10—Earpieces; Attachments therefor ; Earphones; Monophonic headphones
- H04R1/1083—Reduction of ambient noise
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/175—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound
- G10K11/178—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R1/00—Details of transducers, loudspeakers or microphones
- H04R1/10—Earpieces; Attachments therefor ; Earphones; Monophonic headphones
- H04R1/1091—Details not provided for in groups H04R1/1008 - H04R1/1083
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K2210/00—Details of active noise control [ANC] covered by G10K11/178 but not provided for in any of its subgroups
- G10K2210/10—Applications
- G10K2210/108—Communication systems, e.g. where useful sound is kept and noise is cancelled
- G10K2210/1081—Earphones, e.g. for telephones, ear protectors or headsets
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R2201/00—Details of transducers, loudspeakers or microphones covered by H04R1/00 but not provided for in any of its subgroups
- H04R2201/10—Details of earpieces, attachments therefor, earphones or monophonic headphones covered by H04R1/10 but not provided for in any of its subgroups
Abstract
The invention relates to an intelligent noise reduction communication earphone, and belongs to the field of noise reduction and electronic equipment. The environment microphone is positioned at the top of the earphone to acquire an environment noise signal; the communication microphone is connected to one side of the earphone through a communication microphone rotating shaft and acquires a voice signal of personnel communication; the signal preprocessing module is used for completing the amplification, filtering and phase reversal of the environmental noise signal and completing the superposition operation with the amplified and filtered voice signal; the signal analysis processing module comprises a signal digital-to-analog conversion module and a microprocessor system, and is used for completing the functions of acquisition, storage, learning modeling, model-based operation, processing, mode selection and control of the digital voice signal after the environmental noise reverse phase signal, the voice amplification filtering signal and the superposition signal of the environmental noise reverse phase signal and the voice amplification filtering signal; and the wireless transmission module completes the noise reduction and then encodes and transmits the voice signal, receives the voice signal of the call personnel, and outputs the decoded voice signal to the headset. The invention can ensure the voice communication quality without being interfered by the environmental noise.
Description
Technical Field
The invention belongs to the field of noise reduction and electronic equipment, and particularly relates to an intelligent noise reduction communication earphone.
Background
The most basic, convenient and most convenient way for people to communicate is to communicate by voice. In communication among tunnel engineering constructors, tank armored vehicle passengers and other personnel, voice signals can be interfered by various noises from external environments, such as noise interference of surrounding environments, noise interference introduced into transmission media, electric noise interference of communication equipment and the like. The noise interference can pollute the useful voice of the receiving end, when the interference reaches the serious degree, the voice can be completely buried in the noise, and the strong noise environment influences the conversation and communication effects. For noise protection, on one hand, people are equipped with noise insulation earmuffs or helmets to perform passive noise reduction, on the other hand, active noise reduction is adopted, a microphone collects external ambient noise, then a system converts the external ambient noise into sound waves with opposite phases, the sound waves are applied to a horn end, and finally the sound heard by human ears is as follows: the environmental noise and the opposite-phase environmental noise are superposed, so that the noise on the sense organ is reduced, and the method can achieve an effective noise reduction effect to a certain extent. However, because the personnel are in different positions, the noise intensity and the phase in the communication are different, and although the noise is cancelled, larger noise exists; in addition, since the speech characteristics of each person are different, the influence of noise is also different.
Disclosure of Invention
Technical problem to be solved
The invention aims to solve the technical problem of how to provide an intelligent noise reduction communication earphone so as to solve the problem that the noise reduction effect of the existing earphone is not ideal.
(II) technical scheme
In order to solve the above technical problems, the present invention provides an intelligent noise reduction earphone for communication, which comprises an ambient microphone, a call microphone, a signal preprocessing module, a signal analyzing and processing module, and a wireless transmission module,
the environment microphone is positioned at the top of the earphone and used for acquiring an environment noise signal;
the communication microphone is connected to one side of the earphone through a communication microphone rotating shaft, the rotating shaft can rotate downwards to cover the lower part of the lip of a user, and the communication microphone acquires voice signals of person communication;
the signal preprocessing module is used for completing the amplification, filtering and phase reversal of the environmental noise signal and completing the superposition operation with the amplified and filtered voice signal;
the signal analysis processing module comprises a signal digital-to-analog conversion module and a microprocessor system, and is used for completing the functions of acquisition, storage, learning modeling, model-based operation, processing, mode selection and control of the digital voice signal after the environmental noise reverse phase signal, the voice amplification filtering signal and the superposition signal of the environmental noise reverse phase signal and the voice amplification filtering signal;
and the wireless transmission module completes the noise reduction and then encodes and transmits the voice signal, receives the voice signal of the call personnel, and outputs the decoded voice signal to the headset.
Furthermore, the communication microphone is connected with the microphone line, the microphone line is externally protected by the metal hose and the plastic, the microphone line can be randomly bent according to needs, and the communication microphone is located at the tail end of the microphone line.
Furthermore, the signal analysis processing module comprises a sampling control module, a database, a machine learning module and a digital signal processing system, and the superposed signals, the environmental noise signals and the voice signals are sampled at different sampling frequencies under the sampling control module and are stored in different databases; secondly, the signals of the database complete the modeling of a machine learning model in a machine learning module, and the sampled signals realize the separation and noise reduction of the signals under corresponding model parameters; then, the noise reduction of the noise signal is further realized through a corresponding digital signal processing system.
Furthermore, the digital signal processing system comprises two modules, namely a filtering module and a weight adaptive adjustment module; the filtering module determines the structure of the digital signal processing system, and the weight adaptive adjusting module adjusts the weight of each input signal vector, wherein the weight is determined by the parameters of the machine learning model.
Further, the machine learning module provides input signals for the digital signal processing system, provides error signals generated in expected response and adjustment, superposes weighted delay operators through an adder, and adjusts the weight according to model parameters to obtain optimal response signals.
Further, the machine learning module takes the preprocessed noise and voice signals as input of a machine learning model, takes the voice signals obtained in a low-noise environment as output, establishes a training model through machine learning based on the maximum signal-to-noise ratio as a convergence criterion of the model, and outputs model parameters, expected response and error signals generated in adjustment to the digital signal processing system.
Further, the preprocessed noise is ambient noise or white noise.
Further, the machine learning module firstly obtains a large number of voice signals received by the call microphone in a low-noise environment and noise signals in different environments in advance, and superposes the voice signals and the noise signals into noise-containing signals to serve as input signals of machine learning, and takes the low-noise environment signals as output signals to train a noise reduction model; predicting and controlling signals from noisy speech signals requires training to obtain two models: one is a noise signal model and one is a speech signal model; pre-emphasis, framing, windowing, Fast Fourier Transform (FFT), filter bank and Discrete Cosine Transform (DCT) are carried out on the signal to obtain a characteristic coefficient of the signal; after the signal characteristics are extracted, judging whether each frame of signal is a voice signal or not by using a naive Bayes algorithm in machine learning through the characteristics, training and distinguishing a noise signal and a voice signal through a large amount of data and rejecting the noise signal, comparing the signal without the noise with the voice signal in a low-noise environment, judging whether the voice signal without the noise is qualified or not, specifically presetting a signal-to-noise ratio, continuing the next step when the preset signal-to-noise ratio is reached, and continuing the step of rejecting the noise if the preset signal-to-noise ratio is not reached until the preset signal-to-noise ratio reaches the requirements, thereby obtaining a noise reduction model which meets the best signal-to-noise ratio.
Further, the machine learning module reads in the optimal model parameters obtained by machine learning based on the noise reduction model meeting the signal-to-noise ratio, inputs the collected signals to be reduced in noise, and sends the signals to each end of the call through the transmission module after the signals are subjected to machine learning processing.
Furthermore, each communication party generates a noise reduction model of the communication party and outputs a voice signal with the best signal-to-noise ratio.
(III) advantageous effects
The invention provides an intelligent noise reduction communication earphone, which can learn according to noise characteristics and personal voice characteristics, automatically generate a noise reduction model related to noise and personal voice characteristics, optimize noise reduction and send call voice to the other party more clearly; the received voice signal is not interfered by environmental noise, and the quality of voice communication is ensured.
The invention adopts the artificial intelligence principle, models the human voice signal and the environmental noise signal in different environments together, and can realize the optimal signal-to-noise ratio of the voice communication signal in different noise environments. The method has the following specific advantages:
a self-learning function of the communication headset;
outputting and transmitting voice signals under the optimal signal-to-noise ratio criterion based on an artificial intelligence method;
modeling debugging verification in a laboratory environment and intelligent information in a noise working environment.
Drawings
FIG. 1 is a block diagram of the system components of the present invention;
FIG. 2 is a flow chart of machine learning according to the present invention;
FIG. 3 is a block diagram of a digital signal processing system according to the present invention;
FIG. 4 is a flow chart of the machine training of the present invention;
FIG. 5 is a flow chart of the present invention for machine learning in an actual noisy environment;
FIG. 6 is a flow chart of the noise reduction process of the present invention;
FIG. 7 is a white noise based machine learning modeling of the present invention.
Detailed Description
In order to make the objects, contents and advantages of the present invention clearer, the following detailed description of the embodiments of the present invention will be made in conjunction with the accompanying drawings and examples.
Aiming at the problems, the invention provides an intelligent noise reduction method based on artificial intelligence. The method can learn according to the noise characteristics and the individual voice characteristics, automatically generate the noise reduction model related to the noise and the individual voice characteristics, and achieve the purpose of optimizing noise reduction. The conversation voice is sent to the other party more clearly; the received voice signal is not interfered by environmental noise, and the quality of voice communication is ensured.
According to the method, the speech definition evaluation is taken as a criterion, signals acquired by an environmental noise receiving microphone and signals received by a conversation microphone are taken as input signals of a system, speech of a conversation person in a low-noise environment is taken as output of the system, and a machine learning model is established by utilizing a machine learning method. When the voice training system is used, the communication personnel train the system according to respective communication contents, so that respective voices among the communication personnel are clear.
After training is completed, the training can be applied. The invention includes the following aspects:
1. a noise reduction earphone based on artificial intelligence.
Fig. 1 is a block diagram of a system composition. The noise reduction earphone comprises an environment microphone, a call microphone, a signal preprocessing module, a signal analysis processing module and a wireless transmission module.
The ambient microphone is located at the top of the headset and acquires ambient noise signals.
The conversation microphone is connected to one side of the earphone through a conversation microphone rotating shaft, the rotating shaft can rotate downwards and can cover the lower part of the lip of a user, the conversation microphone is connected with a microphone line, the microphone line is externally protected by a metal hose and plastic and can be randomly bent according to needs, and the conversation microphone is positioned at the tail end of the microphone line and is used for acquiring voice signals of personnel communication.
The signal preprocessing module is used for completing the amplification, filtering and phase reversal of the environmental noise signal and completing the superposition operation with the amplified and filtered voice signal;
the signal analysis processing module is composed of a signal digital-to-analog conversion module and a microprocessor system, and is used for completing the functions of acquisition, storage, learning modeling, model-based operation-based digital voice signal processing, mode selection, control and the like of an environmental noise reversed-phase signal, a voice amplification filtering signal and a superposed signal of the environmental noise reversed-phase signal and the voice amplification filtering signal;
and the wireless transmission module completes the noise reduction and then encodes and transmits the voice signal, receives the voice signal of the call personnel, and outputs the decoded voice signal to the headset.
Fig. 2 is a system workflow. Firstly, the system respectively obtains an environmental noise signal and a voice signal from an environmental microphone and a call microphone, the two signals are superposed after different preprocessing, and part of noise of the superposed signals is cancelled. The basic principle of cancellation: let the call microphone be A, the environment microphone be B, A, B are two microphones with the same performance, when the speech call is normal, the mouth is close to the microphone A, it produces larger audio signal Va, at the same time, the microphone B will get some speech signal Vb, but it is much smaller than A, these two signals are inputted into the microphone processor, the two signals are amplified and subtracted, then the obtained signal is Vma＝Vaa-Vba. If there is background noise in the use environment, the intensity of the sound waves reaching the two microphones of the earphone is almost the same, i.e. Va, because the noise source is far from the earphonen≈VbnThus, for background noise, both microphones pick up Vm, althoughn＝Van-VbnAnd the signal is approximately equal to 0, so that the peripheral environmental noise interference can be effectively resisted, and the definition of normal conversation is greatly improved. Ideally, the ambient noise of the call is completely eliminated, but under strong ambient noise conditions, a part of the noise remains in the speech signal, and further noise reduction is required.
As shown in fig. 2, the signal analyzing and processing module includes a sampling control module, a database, a machine learning module and a digital signal processing system, and the superimposed signal, the environmental noise signal and the voice signal are sampled at different sampling frequencies under the sampling control module and stored in different databases. Secondly, the signals of the database complete the modeling of a machine learning model in a machine learning module, and the sampled signals realize the separation and noise reduction of the signals under corresponding model parameters; then, the noise reduction of the noise signal is further realized through a corresponding digital signal processing system.
Fig. 3 is a block diagram of a digital signal processing system. The adaptive filtering device comprises two modules, namely a filtering module and a weight value adaptive adjusting module. The filtering module determines the structure of the digital signal processing system, and the weight adaptive adjusting module adjusts the weight of each input signal vector, wherein the weight is determined by the parameters of the machine learning model. Besides the input signal, a machine learning model is needed to provide an expected response and an error signal generated in the adjustment, weighted delay operators are superposed through an adder, and then the magnitude of the weight is adjusted according to model parameters, so that an optimal response signal can be obtained.
And finally, the codes are sent to other communication personnel at each end. The voice signals of other communication personnel enter the receiving system through the receiving antenna to be demodulated and then are decoded and output to the headset.
2. A noise reduction method based on machine learning.
The machine learning module takes the preprocessed noise (environmental noise or white noise) and voice signals as the input of a machine learning model, takes the voice signals obtained in a low-noise environment as the output, establishes a training model based on the maximum signal-to-noise ratio as the convergence criterion of the model through machine learning, and outputs model parameters, expected response and error signals generated in adjustment to the digital signal processing system. The specific steps are shown in fig. 4, and fig. 5 is a machine learning flowchart.
Firstly, a large number of voice signals received by a call microphone in a low-noise environment and noise signals in different environments are superposed to form a noise-containing signal which is used as an input signal of machine learning, the low-noise environment signal is used as an output signal, and a noise reduction model is trained. Predicting and controlling signals from noisy speech signals requires training to obtain two models: one is a noise signal model and one is a speech signal model. The characteristic coefficients of the signals are obtained after the signals are subjected to the steps of pre-emphasis, framing, windowing, Fast Fourier Transform (FFT), filter bank and Discrete Cosine Transform (DCT). After the signal characteristics are extracted, judging whether each frame of signal is a voice signal or not by using a naive Bayes algorithm in machine learning through the characteristics, training and distinguishing a noise signal and a voice signal through a large amount of data and rejecting the noise signal, comparing the signal without the noise with the voice signal in a low-noise environment, judging whether the voice signal without the noise is qualified or not, specifically presetting a signal-to-noise ratio, continuing the next step when the preset signal-to-noise ratio is reached, and continuing the step of rejecting the noise if the preset signal-to-noise ratio is not reached until the preset signal-to-noise ratio reaches the requirements, thereby obtaining a noise reduction model which meets the best signal-to-noise ratio.
Based on the noise reduction model meeting the best signal-to-noise ratio, the signal after digital-to-analog conversion is input into the model, and the speech signal output with the highest signal-to-noise ratio can be obtained. All communication parties can generate own noise reduction models according to the process and output voice signals with the optimal signal-to-noise ratio. Fig. 6 is a flow chart of the noise reduction operation process. Firstly, reading in optimal model parameters obtained by machine learning, inputting collected signals to be denoised, carrying out machine learning processing, and then sending the signals to each end of a call through a transmission module.
3. Signal transmission
The digital voice signal output by the system is encoded and transmitted in a wired or wireless way, after the opposite side receives the signal, the signal is decoded and then is directly output to the earphone of the receiver after digital-to-analog conversion, and the opposite side communication voice with the highest signal-to-noise ratio can be obtained.
4. Simulation experiment method
In a laboratory environment, an actual noise environment is often lacked for machine learning, and the invention provides a system learning method based on white noise for modeling a system and performing laboratory examination and debugging. According to the method, white noise is used as a noise signal, the white noise signals with different intensities and a low-noise voice signal are superposed to be used as the input of a machine learning system, and the low-noise voice signal is used as the output of the machine learning system to perform machine learning modeling. The method and procedure are the same as 2. The training flow is shown in fig. 7. Firstly, reading in a low-noise voice signal, inputting white noise with different intensities according to the requirement of examination and debugging to simulate the actual noise environment, comparing the low-noise voice signal after machine learning processing with the read-in low-noise voice signal, and judging whether the signal-to-noise ratio is satisfied. And solidifying the model meeting the conditions, storing the parameters and then finishing modeling.
THE ADVANTAGES OF THE PRESENT INVENTION
The invention adopts the artificial intelligence principle, models the human voice signal and the environmental noise signal in different environments together, and can realize the optimal signal-to-noise ratio of the voice communication signal in different noise environments. The method has the following specific advantages:
1. a self-learning function of the communication headset;
2. and outputting and transmitting the voice signal under the optimal signal-to-noise ratio criterion based on an artificial intelligence method.
Modeling debugging verification in a laboratory environment and intelligent information in a noise working environment.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (10)
1. An intelligent noise reduction communication earphone is characterized in that the noise reduction earphone comprises an environment microphone, a call microphone, a signal preprocessing module, a signal analysis processing module and a wireless transmission module,
the environment microphone is positioned at the top of the earphone and used for acquiring an environment noise signal;
the communication microphone is connected to one side of the earphone through a communication microphone rotating shaft, the rotating shaft can rotate downwards to cover the lower part of the lip of a user, and the communication microphone acquires voice signals of person communication;
the signal preprocessing module is used for completing the amplification, filtering and phase reversal of the environmental noise signal and completing the superposition operation with the amplified and filtered voice signal;
the signal analysis processing module comprises a signal digital-to-analog conversion module and a microprocessor system, and is used for completing the functions of acquisition, storage, learning modeling, model-based operation, processing, mode selection and control of the digital voice signal after the environmental noise reverse phase signal, the voice amplification filtering signal and the superposition signal of the environmental noise reverse phase signal and the voice amplification filtering signal;
and the wireless transmission module completes the noise reduction and then encodes and transmits the voice signal, receives the voice signal of the call personnel, and outputs the decoded voice signal to the headset.
2. The intelligent noise reduction communication headset of claim 1, wherein the communication microphone is connected to a microphone line, the microphone line is externally protected by a metal hose and plastic, the microphone line can be bent at will as required, and the communication microphone is located at the tail end of the microphone line.
3. The intelligent noise reduction communication headset according to claim 1, wherein the signal analysis processing module comprises a sampling control module, a database, a machine learning module and a digital signal processing system, and the superimposed signal, the environmental noise signal and the voice signal are sampled at different sampling frequencies under the sampling control module and stored in different databases; secondly, the signals of the database complete the modeling of a machine learning model in a machine learning module, and the sampled signals realize the separation and noise reduction of the signals under corresponding model parameters; then, the noise reduction of the noise signal is further realized through a corresponding digital signal processing system.
4. The intelligent noise reduction communication headset of claim 3, wherein the digital signal processing system comprises two major modules, namely a filtering module and a weight adaptive adjustment module; the filtering module determines the structure of the digital signal processing system, and the weight adaptive adjusting module adjusts the weight of each input signal vector, wherein the weight is determined by the parameters of the machine learning model.
5. The intelligent noise-reducing communication headset of claim 4, wherein the machine learning module provides input signals to the digital signal processing system, provides error signals generated in expected response and adjustment, superimposes weighted delay operators through an adder, and adjusts the weight according to model parameters to obtain an optimal response signal.
6. The intelligent noise reduction communication headset of any one of claims 3-5, wherein the machine learning module takes the preprocessed noise and speech signals as inputs of a machine learning model, takes the speech signals obtained in a low noise environment as outputs, establishes a training model through machine learning based on the maximum signal-to-noise ratio as the convergence criterion of the model, and outputs model parameters, expected response, and error signals generated in adjustment to the digital signal processing system.
7. The intelligent noise-reducing communications headset of claim 6, wherein the pre-processed noise is ambient noise or white noise.
8. The intelligent noise reduction communication headset of claim 6, wherein the machine learning module first obtains in advance a large number of voice signals received by the call microphone in a low noise environment and noise signals in different environments to be superimposed into a noise-containing signal as an input signal for machine learning, and trains a noise reduction model by using the low noise environment signal as an output signal; predicting and controlling signals from noisy speech signals requires training to obtain two models: one is a noise signal model and one is a speech signal model; pre-emphasis, framing, windowing, Fast Fourier Transform (FFT), filter bank and Discrete Cosine Transform (DCT) are carried out on the signal to obtain a characteristic coefficient of the signal; after the signal characteristics are extracted, judging whether each frame of signal is a voice signal or not by using a naive Bayes algorithm in machine learning through the characteristics, training and distinguishing a noise signal and a voice signal through a large amount of data and rejecting the noise signal, comparing the signal without the noise with the voice signal in a low-noise environment, judging whether the voice signal without the noise is qualified or not, specifically presetting a signal-to-noise ratio, continuing the next step when the preset signal-to-noise ratio is reached, and continuing the step of rejecting the noise if the preset signal-to-noise ratio is not reached until the preset signal-to-noise ratio reaches the requirements, thereby obtaining a noise reduction model which meets the best signal-to-noise ratio.
9. The intelligent noise reduction communication headset of claim 8, wherein the machine learning module reads in optimal model parameters obtained by machine learning based on the noise reduction model satisfying the best signal-to-noise ratio, inputs the acquired signal to be noise reduced, and sends the signal to each end of the call through the transmission module after performing machine learning processing.
10. The intelligent noise-reducing communications headset of claim 8, wherein each party to the communication generates its own noise-reducing model that outputs the speech signal with the best signal-to-noise ratio.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111260119.9A CN114007157A (en) | 2021-10-28 | 2021-10-28 | Intelligent noise reduction communication earphone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111260119.9A CN114007157A (en) | 2021-10-28 | 2021-10-28 | Intelligent noise reduction communication earphone |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114007157A true CN114007157A (en) | 2022-02-01 |
Family
ID=79924609
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111260119.9A Pending CN114007157A (en) | 2021-10-28 | 2021-10-28 | Intelligent noise reduction communication earphone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114007157A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114664322A (en) * | 2022-05-23 | 2022-06-24 | 深圳市听多多科技有限公司 | Single-microphone hearing-aid noise reduction method based on Bluetooth headset chip and Bluetooth headset |
| CN114845231A (en) * | 2022-03-25 | 2022-08-02 | 东莞市天翼通讯电子有限公司 | Method and system for testing noise reduction effect of ENC (electronic noise control) through electroacoustic testing equipment |
| CN116633378A (en) * | 2023-07-21 | 2023-08-22 | 江西红声技术有限公司 | Array type voice communication system in helmet |
Citations (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080260180A1 (en) * | 2007-04-13 | 2008-10-23 | Personics Holdings Inc. | Method and device for voice operated control |
| CN101873363A (en) * | 2010-06-09 | 2010-10-27 | 中兴通讯股份有限公司 | Method and terminal for restraining noise by using double microphones |
| US20120084084A1 (en) * | 2010-10-04 | 2012-04-05 | LI Creative Technologies, Inc. | Noise cancellation device for communications in high noise environments |
| CN102411936A (en) * | 2010-11-25 | 2012-04-11 | 歌尔声学股份有限公司 | Speech enhancement method and device as well as head de-noising communication earphone |
| WO2013020380A1 (en) * | 2011-08-10 | 2013-02-14 | 歌尔声学股份有限公司 | Communication headset speech enhancement method and device, and noise reduction communication headset |
| WO2013033946A1 (en) * | 2011-09-10 | 2013-03-14 | 歌尔声学股份有限公司 | Method and system for noise reduction, smart control method and device, and communication device |
| JP2013065039A (en) * | 2012-12-07 | 2013-04-11 | Sony Corp | Headphone, headphone noise reduction method and program for noise reduction processing |
| WO2014187332A1 (en) * | 2013-05-22 | 2014-11-27 | 歌尔声学股份有限公司 | Headset communication method under loud-noise environment and headset |
| CN104602155A (en) * | 2015-01-14 | 2015-05-06 | 中山市天键电声有限公司 | Wireless noise reduction earphone based on intelligent mobile terminal |
| US20150172814A1 (en) * | 2013-12-17 | 2015-06-18 | Personics Holdings, Inc. | Method and system for directional enhancement of sound using small microphone arrays |
| CN104754436A (en) * | 2015-03-13 | 2015-07-01 | 钰太芯微电子科技（上海）有限公司 | Active noise reducing method and noise reducing earphone |
| WO2015135201A1 (en) * | 2014-03-14 | 2015-09-17 | 华为终端有限公司 | Dual-microphone earphone and noise reduction processing method for audio signal in call |
| US20160111009A1 (en) * | 2014-10-17 | 2016-04-21 | Shenzhen Shishangdeyuan Cultural Diffusion Co., Ltd. | Audio noise reduction circuit, an intelligent terminal using this audio noise reduction circuit as well as a teaching method |
| CN105959842A (en) * | 2016-04-29 | 2016-09-21 | 歌尔股份有限公司 | Earphone noise reduction processing method and device, and earphone |
| CN107371079A (en) * | 2017-04-17 | 2017-11-21 | 恒玄科技（上海）有限公司 | The diamylose gram noise reduction system and noise-reduction method of a kind of earphone |
| WO2017206488A1 (en) * | 2016-05-30 | 2017-12-07 | 歌尔股份有限公司 | Noise reduction method and apparatus for self-adaptively adjusting noise reduction gain, and noise reduction earphone |
| CN108712703A (en) * | 2018-03-22 | 2018-10-26 | 恒玄科技（上海）有限公司 | The high-efficient noise-reducing earphone and noise reduction system of low-power consumption |
| CN108831440A (en) * | 2018-04-24 | 2018-11-16 | 中国地质大学（武汉） | A kind of vocal print noise-reduction method and system based on machine learning and deep learning |
| CN109119093A (en) * | 2018-10-30 | 2019-01-01 | Oppo广东移动通信有限公司 | Voice de-noising method, device, storage medium and mobile terminal |
| CN110708625A (en) * | 2019-09-25 | 2020-01-17 | 华东师范大学 | Intelligent terminal-based environment sound suppression and enhancement adjustable earphone system and method |
| CN110769111A (en) * | 2019-10-28 | 2020-02-07 | 珠海格力电器股份有限公司 | Noise reduction method, system, storage medium and terminal |
| CN210868121U (en) * | 2020-02-25 | 2020-06-26 | 广州疯酷科技有限公司 | TWS bluetooth headset of making an uproar falls in two microphones |
| CN111491228A (en) * | 2019-01-29 | 2020-08-04 | 安克创新科技股份有限公司 | Noise reduction earphone and control method thereof |
| CN112053701A (en) * | 2020-09-11 | 2020-12-08 | 北京小米移动软件有限公司 | Sound pickup control method, sound pickup control apparatus, sound pickup control system, sound pickup device, and sound pickup medium |
| CN112312255A (en) * | 2020-06-02 | 2021-02-02 | 深圳市逸音科技有限公司 | Intelligent voice-reducing earphone and noise reduction algorithm based on audio decoder |
| CN113411710A (en) * | 2021-06-29 | 2021-09-17 | 紫优科技(深圳)有限公司 | Voice noise reduction method and device based on double-earphone cooperation |
| CN113436640A (en) * | 2021-06-28 | 2021-09-24 | 歌尔科技有限公司 | Audio noise reduction method, device and system and computer readable storage medium |
| CN214381328U (en) * | 2021-01-26 | 2021-10-08 | 东莞致力电子有限公司 | Earphone of making an uproar falls in conversation |
-
2021
- 2021-10-28 CN CN202111260119.9A patent/CN114007157A/en active Pending
Patent Citations (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080260180A1 (en) * | 2007-04-13 | 2008-10-23 | Personics Holdings Inc. | Method and device for voice operated control |
| CN101873363A (en) * | 2010-06-09 | 2010-10-27 | 中兴通讯股份有限公司 | Method and terminal for restraining noise by using double microphones |
| US20120084084A1 (en) * | 2010-10-04 | 2012-04-05 | LI Creative Technologies, Inc. | Noise cancellation device for communications in high noise environments |
| CN102411936A (en) * | 2010-11-25 | 2012-04-11 | 歌尔声学股份有限公司 | Speech enhancement method and device as well as head de-noising communication earphone |
| WO2013020380A1 (en) * | 2011-08-10 | 2013-02-14 | 歌尔声学股份有限公司 | Communication headset speech enhancement method and device, and noise reduction communication headset |
| WO2013033946A1 (en) * | 2011-09-10 | 2013-03-14 | 歌尔声学股份有限公司 | Method and system for noise reduction, smart control method and device, and communication device |
| JP2013065039A (en) * | 2012-12-07 | 2013-04-11 | Sony Corp | Headphone, headphone noise reduction method and program for noise reduction processing |
| WO2014187332A1 (en) * | 2013-05-22 | 2014-11-27 | 歌尔声学股份有限公司 | Headset communication method under loud-noise environment and headset |
| US20150172814A1 (en) * | 2013-12-17 | 2015-06-18 | Personics Holdings, Inc. | Method and system for directional enhancement of sound using small microphone arrays |
| WO2015135201A1 (en) * | 2014-03-14 | 2015-09-17 | 华为终端有限公司 | Dual-microphone earphone and noise reduction processing method for audio signal in call |
| US20160111009A1 (en) * | 2014-10-17 | 2016-04-21 | Shenzhen Shishangdeyuan Cultural Diffusion Co., Ltd. | Audio noise reduction circuit, an intelligent terminal using this audio noise reduction circuit as well as a teaching method |
| CN104602155A (en) * | 2015-01-14 | 2015-05-06 | 中山市天键电声有限公司 | Wireless noise reduction earphone based on intelligent mobile terminal |
| CN104754436A (en) * | 2015-03-13 | 2015-07-01 | 钰太芯微电子科技（上海）有限公司 | Active noise reducing method and noise reducing earphone |
| CN105959842A (en) * | 2016-04-29 | 2016-09-21 | 歌尔股份有限公司 | Earphone noise reduction processing method and device, and earphone |
| WO2017206488A1 (en) * | 2016-05-30 | 2017-12-07 | 歌尔股份有限公司 | Noise reduction method and apparatus for self-adaptively adjusting noise reduction gain, and noise reduction earphone |
| CN107371079A (en) * | 2017-04-17 | 2017-11-21 | 恒玄科技（上海）有限公司 | The diamylose gram noise reduction system and noise-reduction method of a kind of earphone |
| CN108712703A (en) * | 2018-03-22 | 2018-10-26 | 恒玄科技（上海）有限公司 | The high-efficient noise-reducing earphone and noise reduction system of low-power consumption |
| CN108831440A (en) * | 2018-04-24 | 2018-11-16 | 中国地质大学（武汉） | A kind of vocal print noise-reduction method and system based on machine learning and deep learning |
| CN109119093A (en) * | 2018-10-30 | 2019-01-01 | Oppo广东移动通信有限公司 | Voice de-noising method, device, storage medium and mobile terminal |
| CN111491228A (en) * | 2019-01-29 | 2020-08-04 | 安克创新科技股份有限公司 | Noise reduction earphone and control method thereof |
| CN110708625A (en) * | 2019-09-25 | 2020-01-17 | 华东师范大学 | Intelligent terminal-based environment sound suppression and enhancement adjustable earphone system and method |
| CN110769111A (en) * | 2019-10-28 | 2020-02-07 | 珠海格力电器股份有限公司 | Noise reduction method, system, storage medium and terminal |
| CN210868121U (en) * | 2020-02-25 | 2020-06-26 | 广州疯酷科技有限公司 | TWS bluetooth headset of making an uproar falls in two microphones |
| CN112312255A (en) * | 2020-06-02 | 2021-02-02 | 深圳市逸音科技有限公司 | Intelligent voice-reducing earphone and noise reduction algorithm based on audio decoder |
| CN112053701A (en) * | 2020-09-11 | 2020-12-08 | 北京小米移动软件有限公司 | Sound pickup control method, sound pickup control apparatus, sound pickup control system, sound pickup device, and sound pickup medium |
| CN214381328U (en) * | 2021-01-26 | 2021-10-08 | 东莞致力电子有限公司 | Earphone of making an uproar falls in conversation |
| CN113436640A (en) * | 2021-06-28 | 2021-09-24 | 歌尔科技有限公司 | Audio noise reduction method, device and system and computer readable storage medium |
| CN113411710A (en) * | 2021-06-29 | 2021-09-17 | 紫优科技(深圳)有限公司 | Voice noise reduction method and device based on double-earphone cooperation |
Non-Patent Citations (3)
| Title |
|---|
| 樊伟;曾庆宁;: "一种基于麦克风阵列的宽带语音增强算法研究", 大众科技, no. 12 * |
| 樊哲宇;赵航;李美敏;次仁卓玛;: "噪声环境下使用主动降噪耳机对听力的保护效果研究", 中国听力语言康复科学杂志, no. 02 * |
| 葛艳: "《低信噪比下的语音端点检测技术的研究》", 《硕士电子期刊》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114845231A (en) * | 2022-03-25 | 2022-08-02 | 东莞市天翼通讯电子有限公司 | Method and system for testing noise reduction effect of ENC (electronic noise control) through electroacoustic testing equipment |
| CN114845231B (en) * | 2022-03-25 | 2023-01-24 | 东莞市天翼通讯电子有限公司 | Method and system for testing noise reduction effect of ENC (electronic noise control) through electroacoustic testing equipment |
| CN114664322A (en) * | 2022-05-23 | 2022-06-24 | 深圳市听多多科技有限公司 | Single-microphone hearing-aid noise reduction method based on Bluetooth headset chip and Bluetooth headset |
| CN114664322B (en) * | 2022-05-23 | 2022-08-12 | 深圳市听多多科技有限公司 | Single-microphone hearing-aid noise reduction method based on Bluetooth headset chip and Bluetooth headset |
| CN116633378A (en) * | 2023-07-21 | 2023-08-22 | 江西红声技术有限公司 | Array type voice communication system in helmet |
| CN116633378B (en) * | 2023-07-21 | 2023-12-08 | 江西红声技术有限公司 | Array type voice communication system in helmet |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114007157A (en) | Intelligent noise reduction communication earphone | |
| US9812147B2 (en) | System and method for generating an audio signal representing the speech of a user | |
| CN101222555B (en) | System and method for improving audio speech quality | |
| CN111885275B (en) | Echo cancellation method and device for voice signal, storage medium and electronic device | |
| CN107993670A (en) | Microphone array voice enhancement method based on statistical model | |
| US20230352038A1 (en) | Voice activation detecting method of earphones, earphones and storage medium | |
| KR101433833B1 (en) | Method and System for Providing an Acoustic Signal with Extended Bandwidth | |
| CN105448302B (en) | A kind of the speech reverberation removing method and system of environment self-adaption | |
| JP2010055024A (en) | Signal correction device | |
| CN113241085B (en) | Echo cancellation method, device, equipment and readable storage medium | |
| CN110383798A (en) | Acoustic signal processing device, acoustics signal processing method and hands-free message equipment | |
| CN110310653A (en) | A kind of echo cancel method | |
| CN108022595A (en) | A kind of voice signal noise-reduction method and user terminal | |
| KR102307355B1 (en) | Method and apparatus for improving call quality in noisy environment | |
| CN114792524B (en) | Audio data processing method, apparatus, program product, computer device and medium | |
| CN113823304A (en) | Voice signal processing method and device, electronic equipment and readable storage medium | |
| CN112259112A (en) | Echo cancellation method combining voiceprint recognition and deep learning | |
| CN108133712A (en) | A kind of method and apparatus for handling audio data | |
| CN114338623B (en) | Audio processing method, device, equipment and medium | |
| CN113782044A (en) | Voice enhancement method and device | |
| CN116030823A (en) | Voice signal processing method and device, computer equipment and storage medium | |
| CN109215635B (en) | Broadband voice frequency spectrum gradient characteristic parameter reconstruction method for voice definition enhancement | |
| CN112489669B (en) | Audio signal processing method, device, equipment and medium | |
| CN111968627B (en) | Bone conduction voice enhancement method based on joint dictionary learning and sparse representation | |
| CN102054482B (en) | Method and device for enhancing 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 |