CN109451386A - Return sound functional component, sound insulation feedback earphone and its application and sound insulation feedback method - Google Patents
Return sound functional component, sound insulation feedback earphone and its application and sound insulation feedback method Download PDFInfo
- Publication number
- CN109451386A CN109451386A CN201811225396.4A CN201811225396A CN109451386A CN 109451386 A CN109451386 A CN 109451386A CN 201811225396 A CN201811225396 A CN 201811225396A CN 109451386 A CN109451386 A CN 109451386A
- Authority
- CN
- China
- Prior art keywords
- sound
- earphone
- sound insulation
- conversion module
- functional component
- 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
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
- H04R1/1091—Details not provided for in groups H04R1/1008 - H04R1/1083
-
- 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 discloses one kind to return sound functional component, sound insulation feedback earphone and its application and sound insulation feedback method, and the sound functional component includes that sequentially connected single directive microphone receives head (1), sound intensifier (2), AD conversion module (3), processor (4), D/A conversion module (5) and audio amplifier (6).The sound that user itself is only returned under noisy environment may be implemented in the present invention;In addition, by the way that the sound functional component that returns of the invention is loaded into common earphone, so as to realize multiple students in same classroom while practicing spoken language or chant, it avoids mutually being interfered, high frequency, low frequency noise can be filtered out simultaneously, the sound for only returning to user itself not will cause audio distortions, and require the composition material of wearing mode and earphone lower;It can seek peace and quiet in noisy surroundings at any time, improve verbal learning efficiency.
Description
Technical field
The present invention relates to one kind to return sound functional component, sound insulation feedback earphone and its application and sound insulation feedback method, belongs to ear
Machine technical field.
Background technique
When multiple students practice spoken language or chant simultaneously in same classroom, it will cause and interfere with each other and influence study effect
Fruit.Sound isolating earphone or earplug are put on, and has the feeling of distortion.In spoken self-learning process, compareed with standard pronunciation and intelligent
The mistake of voice is corrected, it is also very crucial, but can not allow teacher's moment that can be concerned about all students.Therefore, with Intellgent Ear
Machine solves problems, there is certain realistic meaning.
In existing earphone noise reduction technology, it is broadly divided into active noise reduction and physical isolation two schemes.The original of active noise reduction
Reason is to sample by a small microphone to ambient noise using signal processing circuit, a reverse phase is then generated in earphone
Signal to eliminate noise.Due to utilizing small microphone pickup, it is impossible to sample 100% signal, cannot also export
100% accurate inversion signal will cause certain distortion, while to the voice signal of user itself also without special place
Reason, can filter off together.The passive noise reduction of earphone, that is, with the method for physics sound insulation, the sound outside earphone is separated as far as possible, is only allowed
Ear hears the sound in earphone.This method, it is desirable that earphone surrounds ear and forms enclosure space, to wearing mode and earphone
Composition material is more demanding, and can only also obstruct high-frequency noise, unobvious to low-frequency noise effect.
The earphone class product both at home and abroad with sound insulation feedback and voice error correction is all to use traditional filtering at present
(a kind of such as entitled earphone that can hear oneself sound, application No. is 201210116161.8 applications for a patent for invention for scheme;Name
Referred to as K sings/sings that synchronous monitoring headpone, application No. is 201610522760.8 applications for a patent for invention;A kind of entitled noise reduction
The measurement method of earphone, application No. is 201810167336.5 applications for a patent for invention), it is all mainly benefit during returning sound
With the scheme of correct sound and cancel with noise or frequency filtering, the sound of user itself can not be only returned under noisy environment
Sound, and correct pronunciation can only be played, carry out error correction can not be read aloud to user, be suitable for assisting Foreigh-language oral-speech there are no one
Trained earphone.
In addition, at this stage, speech recognition technology has been achieved for substantive breakthrough, many successful speech recognition systems
It comes out one after another.However the evaluation and test of these identifying system performances is realized under opposite quiet environment, when in a noisy environment
When use, their performance can sharply decline, or even can not identify.Currently used recognition methods mainly has dynamic time warping
(dynamic time warping, DTW) method, vector quantization (vector quantization, VQ) method and hidden Ma Erke
Husband's model (hidden Markov model, HMM) method.But with going deep into for research, the limitation of conventional speech recognition methods
Also gradually it is exposed, is mainly difficult to solve following problems:
(1) ambient noise problem.When signal-to-noise ratio decline, identification error rate is doubled and redoubled, and current various anti-noises are calculated
Method, mainly by the filter in terms of some frequency domains, although improving the identity under noise conditions to a certain extent
Can, but error rate under the conditions of high s/n ratio increased.
(2) spoken voice problem.When speaker leaves preset text, inevitably generates unfluent phenomenon and be not inconsistent
Syntax gauge phenomenon is closed, it is full that lamprophonia is usually not read aloud in pronunciation.These phenomenons result in acoustic model and language model
Double mismatch, sharply decline so as to cause discrimination.
(3) more speaker's problems.It, still can without a kind of good algorithm in traditional algorithm when multiple speakers speak simultaneously
To separate lap, also without a kind of support multiple target while decoded speech recognition algorithm.
Summary of the invention
It is an object of the present invention to provide one kind to return sound functional component, may be implemented only to return under noisy environment
The sound at reuse family itself;
It is another object of the present invention to provide a kind of sound insulation feedback earphone and its application and a kind of sound insulation feedback sides
Method is practiced spoken language simultaneously in same classroom for multiple students or is chanted, avoids mutually being interfered, can filter simultaneously
Fall high frequency, low frequency noise, only returns to the sound of user itself, not will cause audio distortions, and to the group of wearing mode and earphone
It is lower at material requirements;In addition;Sound insulation feedback earphone of the invention can read aloud carry out error correction to user.
In order to solve the above technical problems, the present invention adopts the following technical scheme that: one kind returning sound functional component, including sequentially
The single directive microphone of connection receives head, sound intensifier, AD conversion module, processor, D/A conversion module and audio amplifier.
Preferably, the single directive microphone receives single directional microphone that head uses cardioid to be directed toward.The present invention
Voice signal is obtained by selecting cardioid directional microphone to receive head, using closely saying effect possessed by cardioid directional microphone
It answers, effectively removes environment noise, amplify the voice signal of user ratio shared in the signal received, to obtain more
Useful signal, while cardioid be directed toward single directional microphone it is low in cost, collected interference signal is few.
Preferably, the sound intensifier uses MAX9812 chip;The AD conversion module, processor, DA conversion
Module uses MSP430F5244 chip microcontroller;Audio amplifier uses audio amplifier PAM8406.
Sound intensifier in the present invention uses MAX9812 chip, and the output signal of microphone is converted to electric signal, this
Kind chip has package dimension small, built-in low noise microphone biasing, a variety of good characteristics such as energy conservation;Simultaneously in view of earphone is answered
Has easy to carry and continuation of the journey for a long time characteristic, the present invention selects MSP430F5244 single-chip microcontroller as main control chip, because it is same
When have AD and D/A module, so as to reduce peripheral circuit, while the single-chip microcontroller of msp430 series has low-voltage, super low-power consumption
Characteristic, be conducive to continue a journey for a long time;Because the electric current of single-chip microcontroller is not enough to drive the speaker, the present invention has selected number
Word power amplifier chips PAM8406 drive the speaker, it is small in size using its, and have the characteristics of outstanding noise suppressing function, reach
One preferable effect.
A kind of sound insulation feedback earphone, comprising: any one of above-mentioned to return sound functional component and headset body, the earphone sheet
Body includes loudspeaker, and the sound functional component that returns is connect with loudspeaker.
A kind of sound insulation feedback method of sound insulation feedback earphone above-mentioned, comprising the following steps:
S1, single directive microphone receive head acquisition voice signal and are effectively gone using effect is closely said possessed by microphone
Except environment noise;
S2 converts electric signal for the voice signal using sound intensifier, is then carried out using AD conversion module
AD sampling obtains data;
S3, processor are eliminated single directive microphone using ICA independent composition analysis algorithm and are received in head sampling process
Noise, individually export user itself voice signal, then by the speech signal by D/A conversion module conversion after export to
Audio amplifier;
S4 is exported again by amplified signal to the loudspeaker of sound isolating earphone.
Sound insulation feedback method above-mentioned, in step S3, processor is using ICA independent composition analysis algorithm to the sound of acquisition
Signal carries out independent component analysis, obtains the hybrid matrix of signal;Composition voice signal is acquired using the inverse matrix of hybrid matrix
Each component;The voice signal of the maximum as user of mean amplitude of tide in these components.
By isolating various signals from audio signal using ICA algorithm in the present invention, inventor is flat by recycling
Equal amplitude selects sound of the maximum signal of volume as user, so that the processor has the effect of a filter.
Relative to the filter of traditional frequency domain, it can not only filter off the great noise of frequency phase-difference, can also remove it simultaneously
The interference of other people sound, be because of the frequency of the sound of people similar in, can not be filtered off with traditional scheme.
A kind of application of above-mentioned sound insulation feedback earphone carries out Foreigh-language oral-speech auxiliary using the sound insulation feedback earphone and practices
It practises.
The application of sound insulation feedback earphone above-mentioned, comprising the following steps:
A feeds back earphone from sound insulation by bluetooth and obtains voice data;
The voice data is transmitted to speech recognition API, converts thereof into corresponding lteral data, and according to described by b
Lteral data, its corresponding standard pronunciation is obtained from database;
C compares the sound ray that standard pronunciation sound ray is recorded with user, thus obtain user's spoken language there are the problem of, and
It displays it on screen, user is helped to promote oracy faster.
Preferably, step c includes: according to the distributing order of the correct phonetic symbol of word identified and duration by user
Sound be cut into each phonetic symbol, then the phonetic symbol is compared with the phonetic symbol of standard to (some spies provided compared with mainly
Value indicative, such as variance etc);It makes marks on the phonetic symbol if differing larger, reminds user's correction.
Compared with prior art, the invention has the following advantages that
1, by using returned constructed by the present invention sound functional component (including sequentially connected single directive microphone receive
Head, sound intensifier, AD conversion module, processor, D/A conversion module and audio amplifier), so as to realize in noisy environment
The sound of lower return user itself;
2, by the way that the sound functional component that returns of the invention to be loaded into common earphone, Foreigh-language oral-speech is carried out for assisting,
So as to realize multiple students in same classroom while practicing spoken language or chant, avoid mutually being interfered, Ke Yitong
When filter out high frequency, low frequency noise (specific ICA independent composition analysis algorithm (and the unconventional frequency using in machine learning
Domain filters solutions), classify to sound, retain that maximum one kind of classification amplitude, thus from the observation letter of one group of mixing
Independent signal is isolated in number.Such as in a big room, many people are speaking simultaneously, and sample is each position in this room
The Duan Luyin set, ICA can isolate everyone independent one's voice in speech from the recording that these are mixed.As long as can be from solution
In each sound separated out, the sound for finding out user can filter off the interference of other sound, and this sound can be
Belong to any frequency band, include high and low frequency), the sound of user itself is only returned, not will cause audio distortions, and to pendant
The composition material requirement for wearing mode and earphone is lower;It can seek peace and quiet in noisy surroundings at any time, improve verbal learning efficiency.
3, the present invention carries out spoken correction by being accurate to the scheme of phonetic symbol in the spoken correction of auxiliary, so as to more smart
The Oral English Practice of true correction user.
4, earphone is fed back in sound insulation of the invention, due to having there is enough filter effects after returning sound functional component,
The sound isolating earphone that there is no need to use price more expensive again passes to general after need to being only filtered from the sound that audio amplifier comes out
Loudspeaker in logical earphone, can be realized sound insulation and anti-sound function.
5, the present invention obtains voice signal, the principle based on barometric gradient, using closely saying using cardioid directional microphone
It after effect removes most of environment noise, then carries out AD sampling and obtains data, innovative to the data of acquisition is independent using ICA
Constituent analysis algorithm carries out the filtering of software aspects, eliminates the noise in single directive microphone reception head sampling process, thus
Personal sound can be preferably extracted, while possessing preferable anti-interference ability.
6, by feeding back the application of the bluetooth communication between earphone and mobile phone and cell phone application, benefit using sound insulation of the invention
With the CPU, memory and internet function of mobile phone, to realize the voice error correction to declaimer.
7, the invention enables speech recognition performances to be promoted to a completely new level, is carried out using sound insulation feedback earphone auxiliary
Foreigh-language oral-speech practice can effectively solve the problems, such as to will receive interference when people practice spoken under noisy environment.
Although single directive microphone has preferable directive property, external interference still can be inevitably received
Signal, although amplitude is smaller, the sound after audio amplifier can seem more noisy, need to filter out.Inventor is not by
After disconnected trial, ICA independent composition analysis algorithm is had finally chosen, it can find most accurate sound by constantly training
The hybrid matrix of sound can isolate what each sound source was issued from the sound that sampling obtains by the inverse matrix of hybrid matrix
The component of sound, as long as comparing their amplitude again can be obtained use using the maximum sound of amplitude as the sound of user
The sound component at family is filtered the voice signal of collection in worksite in real time to realize, and can accurately be extracted
The sound of user out.
Detailed description of the invention
Fig. 1 is a kind of structural schematic diagram of embodiment of the invention;
Fig. 2 is the flow chart of sound insulation feedback method in the present invention;
Fig. 3 is the application method flow chart of sound insulation feedback earphone in the present invention;
Fig. 4 is the structural schematic diagram of sound insulation feedback earphone.
Appended drawing reference: 1- single directive microphone receives head, 2- sound intensifier, 3-AD conversion module, 4- processor, 5-
D/A conversion module, 6- audio amplifier, 7- headset body, 8- loudspeaker.
The present invention is further illustrated with reference to the accompanying drawings and detailed description.
Specific embodiment
The embodiment of the present invention 1: one kind returning sound functional component, as shown in Figure 1, including sequentially connected unidirectivity Mike
Wind receives head 1, sound intensifier 2, AD conversion module 3, processor 4, D/A conversion module 5 and audio amplifier 6.Single direction
Property microphone receive head 1 using cardioid be directed toward single directional microphone.The sound intensifier 2 uses MAX9812 chip;
The AD conversion module 3, processor 4, D/A conversion module 5 use MSP430F5244 chip microcontroller;Audio amplifier 6 uses
Audio amplifier PAM8406.
Embodiment 2: one kind returning sound functional component, including sequentially connected single directive microphone receives head 1, sound amplification
Device 2, AD conversion module 3, processor 4, D/A conversion module 5 and audio amplifier 6.The single directive microphone receives head 1 and adopts
The single directional microphone being directed toward with cardioid.
Embodiment 3: one kind returning sound functional component, including sequentially connected single directive microphone receives head 1, sound amplification
Device 2, AD conversion module 3, processor 4, D/A conversion module 5 and audio amplifier 6.The sound intensifier 2 uses MAX9812 core
Piece;The AD conversion module 3, processor 4, D/A conversion module 5 use MSP430F5244 chip microcontroller;Audio amplifier 6 is adopted
With audio amplifier PAM8406.The single directive microphone is received head 1 and can be directed toward using super heart-shaped direction, acute core shape, is flat
All kinds of single directional microphones that cardioid is directed toward.
Embodiment 4: one kind returning sound functional component, including sequentially connected single directive microphone receives head 1, sound amplification
Device 2, AD conversion module 3, processor 4, D/A conversion module 5 and audio amplifier 6.The single directive microphone receives head 1 can
Using super heart-shaped direction, all kinds of single directional microphones that acute core shape is directed toward, flat cardioid is directed toward.The sound intensifier 2,
AD conversion module 3, processor 4, D/A conversion module 5 and audio amplifier 6 can be realized using commercially available existing module.
Embodiment 5: a kind of sound insulation feedback earphone, as shown in Figure 4, comprising: return sound function described in embodiment 1 to 4 is any
Component and headset body 7, the headset body 7 include loudspeaker 8, and the sound functional component that returns is connect with loudspeaker 8.
Embodiment 6: sound insulation described in embodiment 5 feed back earphone a kind of sound insulation feedback method, as shown in Fig. 2, include with
Lower step:
S1, single directive microphone receive head 1 and obtain voice signal, using effect is closely said possessed by microphone, effectively
Remove environment noise;
S2 converts electric signal for the voice signal using sound intensifier 2, then using AD conversion module 3 into
Row AD sampling obtains data;
S3, processor 4 are eliminated single directive microphone using ICA independent composition analysis algorithm and are received in 1 sampling process of head
Noise, individually export user itself voice signal, then by the speech signal by D/A conversion module 5 convert after export
To audio amplifier 6;Wherein, processor 4 carries out independent element point using voice signal of the ICA independent composition analysis algorithm to acquisition
Analysis, obtains the hybrid matrix of signal;Each component of composition voice signal is acquired using the inverse matrix of hybrid matrix;These components
The voice signal of the maximum as user of middle mean amplitude of tide;
S4 is exported again by amplified signal to the loudspeaker 8 of sound isolating earphone.
A kind of embodiment 7: sound insulation feedback method of the feedback earphone of sound insulation described in embodiment 5, comprising the following steps:
S1, single directive microphone receive head 1 and obtain voice signal, using effect is closely said possessed by microphone, effectively
Remove environment noise;
S2 converts electric signal for the voice signal using sound intensifier 2, then using AD conversion module 3 into
Row AD sampling obtains data;
S3, processor 4 are eliminated single directive microphone using ICA independent composition analysis algorithm and are received in 1 sampling process of head
Noise, individually export user itself voice signal, then by the speech signal by D/A conversion module 5 convert after export
To audio amplifier 6;
S4 is exported again by amplified signal to the loudspeaker 8 of sound isolating earphone.
In step S3, it can also be decomposed using SVD or various other methods such as PCA principal component analysis eliminate unidirectivity wheat
Gram wind receives the noise in 1 sampling process of head.
Embodiment 8: a kind of application of earphone is fed back in sound insulation described in embodiment 5, as shown in figure 3, utilizing the sound insulation
It feeds back earphone and carries out Foreigh-language oral-speech auxiliary exercise;The following steps are included:
A feeds back earphone from sound insulation by bluetooth and obtains voice data;
The voice data is transmitted to speech recognition API, converts thereof into corresponding lteral data, and according to described by b
Lteral data, its corresponding standard pronunciation is provided from the database (such as database of Baidu's voice, Iflytek offer);
C, the sound ray that standard pronunciation sound ray and user record is compared (for example mark out in each word and standard pronunciation
Differ that maximum several phonetic symbol), thus obtain user's spoken language there are the problem of (can to user broadcast emphatically), and shown
Show on screen, can mark out simultaneously inconsistent reason: that such as drags is too long, volume deficiency etc., and auxiliary user carries out foreign language
Spoken training.
Step c is specifically included: according to the distributing order of the correct phonetic symbol of the word identified and duration by user's
Sound is cut into each phonetic symbol, then the phonetic symbol is compared with the phonetic symbol of standard to (some features provided compared with mainly
Value, such as variance etc);It makes marks on the phonetic symbol if differing larger, reminds user's correction.
Embodiment 9: a kind of application of earphone is fed back in sound insulation described in embodiment 5, using the sound insulation feed back earphone into
Row Foreigh-language oral-speech auxiliary exercise.
Claims (9)
1. one kind returns sound functional component, which is characterized in that receive head (1), sound including sequentially connected single directive microphone
Amplifier (2), AD conversion module (3), processor (4), D/A conversion module (5) and audio amplifier (6).
2. according to claim 1 return sound functional component, which is characterized in that the single directive microphone receives head
(1) the single directional microphone being directed toward using cardioid.
3. according to claim 1 or 2 return sound functional component, which is characterized in that the sound intensifier (2) uses
MAX9812 chip;The AD conversion module (3), processor (4), D/A conversion module (5) use MSP430F5244 single-chip microcontroller
It realizes;Audio amplifier (6) uses audio amplifier PAM8406.
4. earphone is fed back in a kind of sound insulation characterized by comprising claims 1 to 3 is described in any item return sound functional component and
Headset body (7), the headset body (7) include loudspeaker (8), and the sound functional component that returns is connect with loudspeaker (8).
5. a kind of sound insulation feedback method of sound insulation feedback earphone as claimed in claim 4, which comprises the following steps:
S1, single directive microphone receive head (1) acquisition voice signal and are effectively gone using effect is closely said possessed by microphone
Except environment noise;
S2 converts electric signal for the voice signal using sound intensifier (2), then using AD conversion module (3) into
Row AD sampling obtains data;
S3, processor (4) are eliminated single directive microphone using ICA independent composition analysis algorithm and are received in head (1) sampling process
Noise, individually export user itself voice signal, then by the speech signal by D/A conversion module (5) convert after it is defeated
Out to audio amplifier (6);
S4 is exported by amplified signal again to the loudspeaker of sound isolating earphone (9).
6. sound insulation feedback method according to claim 5, which is characterized in that in step S3, processor (4) is only using ICA
Vertical constituent analysis algorithm carries out independent component analysis to the voice signal of acquisition, obtains the hybrid matrix of signal;Utilize mixed moment
The inverse matrix of battle array acquires each component of composition voice signal;The voice letter of the maximum as user of mean amplitude of tide in these components
Number.
7. a kind of application of sound insulation feedback earphone as claimed in claim 4, which is characterized in that feed back earphone using the sound insulation
Carry out Foreigh-language oral-speech auxiliary exercise.
8. the application of sound insulation feedback earphone according to claim 7, which comprises the following steps:
A feeds back earphone from sound insulation by bluetooth and obtains voice data;
The voice data is transmitted to speech recognition API by b, converts thereof into corresponding lteral data, and according to the text
Digital data obtains its corresponding standard pronunciation from database;
C compares the sound ray that standard pronunciation sound ray is recorded with user, thus obtain user's spoken language there are the problem of, and by its
It is shown on screen.
9. the application of sound insulation feedback earphone according to claim 8, which is characterized in that step c includes: that basis identifies
The sound of user is cut into each phonetic symbol by the distributing order of the correct phonetic symbol of word and duration, then by the phonetic symbol with
The phonetic symbol of standard compares;It makes marks on the phonetic symbol if differing larger, reminds user's correction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811225396.4A CN109451386A (en) | 2018-10-20 | 2018-10-20 | Return sound functional component, sound insulation feedback earphone and its application and sound insulation feedback method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811225396.4A CN109451386A (en) | 2018-10-20 | 2018-10-20 | Return sound functional component, sound insulation feedback earphone and its application and sound insulation feedback method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109451386A true CN109451386A (en) | 2019-03-08 |
Family
ID=65547732
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811225396.4A Pending CN109451386A (en) | 2018-10-20 | 2018-10-20 | Return sound functional component, sound insulation feedback earphone and its application and sound insulation feedback method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109451386A (en) |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1512300A (en) * | 2002-12-30 | 2004-07-14 | 艾尔科技股份有限公司 | User's interface, system and method for automatically marking phonetic symbol to correct pronunciation |
CN101031956A (en) * | 2004-07-22 | 2007-09-05 | 索福特迈克斯有限公司 | Headset for separation of speech signals in a noisy environment |
CN101996639A (en) * | 2009-08-12 | 2011-03-30 | 财团法人交大思源基金会 | Audio signal separating device and operation method thereof |
CN202331850U (en) * | 2011-11-25 | 2012-07-11 | 包春梅 | Foreign language learning machine capable of marking phonetic symbol automatically to correct pronunciation |
CN205232421U (en) * | 2015-12-30 | 2016-05-11 | 辽宁对外经贸学院 | A earphone for english listening comprehension and spoken study |
CN106548787A (en) * | 2016-11-01 | 2017-03-29 | 上海语知义信息技术有限公司 | The evaluating method and evaluating system of optimization new word |
CN206249614U (en) * | 2016-10-25 | 2017-06-13 | 齐齐哈尔医学院 | A kind of pronunciation correction system of English Listening Comprehension and spoken equipment |
CN107767862A (en) * | 2017-11-06 | 2018-03-06 | 深圳市领芯者科技有限公司 | Voice data processing method, system and storage medium |
CN209234005U (en) * | 2018-10-20 | 2019-08-09 | 东北大学秦皇岛分校 | It returns sound functional component and assists the sound insulation feedback earphone of Foreigh-language oral-speech training |
-
2018
- 2018-10-20 CN CN201811225396.4A patent/CN109451386A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1512300A (en) * | 2002-12-30 | 2004-07-14 | 艾尔科技股份有限公司 | User's interface, system and method for automatically marking phonetic symbol to correct pronunciation |
CN101031956A (en) * | 2004-07-22 | 2007-09-05 | 索福特迈克斯有限公司 | Headset for separation of speech signals in a noisy environment |
CN101996639A (en) * | 2009-08-12 | 2011-03-30 | 财团法人交大思源基金会 | Audio signal separating device and operation method thereof |
CN202331850U (en) * | 2011-11-25 | 2012-07-11 | 包春梅 | Foreign language learning machine capable of marking phonetic symbol automatically to correct pronunciation |
CN205232421U (en) * | 2015-12-30 | 2016-05-11 | 辽宁对外经贸学院 | A earphone for english listening comprehension and spoken study |
CN206249614U (en) * | 2016-10-25 | 2017-06-13 | 齐齐哈尔医学院 | A kind of pronunciation correction system of English Listening Comprehension and spoken equipment |
CN106548787A (en) * | 2016-11-01 | 2017-03-29 | 上海语知义信息技术有限公司 | The evaluating method and evaluating system of optimization new word |
CN107767862A (en) * | 2017-11-06 | 2018-03-06 | 深圳市领芯者科技有限公司 | Voice data processing method, system and storage medium |
CN209234005U (en) * | 2018-10-20 | 2019-08-09 | 东北大学秦皇岛分校 | It returns sound functional component and assists the sound insulation feedback earphone of Foreigh-language oral-speech training |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Eaton et al. | The ACE challenge—Corpus description and performance evaluation | |
CN101023469B (en) | Digital filtering method, digital filtering equipment | |
CN110970053B (en) | Multichannel speaker-independent voice separation method based on deep clustering | |
CN106782591B (en) | Device and method for improving speech recognition rate under background noise | |
CN110473567A (en) | Audio-frequency processing method, device and storage medium based on deep neural network | |
US10277750B2 (en) | Method and system for improving echo in hands-free call of mobile terminal | |
Moore et al. | Microphone array speech recognition: Experiments on overlapping speech in meetings | |
CN110234051A (en) | A kind of anti-public address method and system of uttering long and high-pitched sounds based on deep learning | |
CN107274887A (en) | Speaker's Further Feature Extraction method based on fusion feature MGFCC | |
CN101447189A (en) | Voice interference method | |
Healy et al. | A talker-independent deep learning algorithm to increase intelligibility for hearing-impaired listeners in reverberant competing talker conditions | |
CN111276150A (en) | Intelligent voice-to-character and simultaneous interpretation system based on microphone array | |
Liang et al. | Attention-based multi-channel speaker verification with ad-hoc microphone arrays | |
CN214226506U (en) | Sound processing circuit, electroacoustic device, and sound processing system | |
CN112349291A (en) | Sound amplification system and method based on AI noise reduction model | |
CN209234005U (en) | It returns sound functional component and assists the sound insulation feedback earphone of Foreigh-language oral-speech training | |
CN109451386A (en) | Return sound functional component, sound insulation feedback earphone and its application and sound insulation feedback method | |
Jaroslavceva et al. | Robot Ego‐Noise Suppression with Labanotation‐Template Subtraction | |
Friedrichs et al. | Vowel recognition at fundamental frequencies up to 1 kHz reveals point vowels as acoustic landmarks | |
CN206181171U (en) | Outer sound reproduction matter automatic regulating system of cell -phone | |
CN114120965A (en) | Audio processing method, electronic device, and storage medium | |
CN114979887A (en) | Wisdom education earphone | |
CN102426839B (en) | Voice recognition method for deaf people | |
Geng et al. | A speech enhancement method based on the combination of microphone array and parabolic reflector | |
CN114189781A (en) | Noise reduction method and system for double-microphone neural network noise reduction earphone |
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 |