CN1896955A - Electronic canails audition simulating platform, its simulating experimental method and algorithm comparation - Google Patents
Electronic canails audition simulating platform, its simulating experimental method and algorithm comparation Download PDFInfo
- Publication number
- CN1896955A CN1896955A CN 200610060579 CN200610060579A CN1896955A CN 1896955 A CN1896955 A CN 1896955A CN 200610060579 CN200610060579 CN 200610060579 CN 200610060579 A CN200610060579 A CN 200610060579A CN 1896955 A CN1896955 A CN 1896955A
- Authority
- CN
- China
- Prior art keywords
- hearing
- sense
- emulation
- processing algorithm
- phonetic
- 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.)
- Granted
Links
Images
Landscapes
- Prostheses (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
An auditory artificial platform of electronic cochlea consists of auditory artificial software platform, program library of auditory artificial mode, voice material databank and auditory artificial test databank. It is featured as not only enabling to predict out electric-auditory result of user applying electronic cochlea with different voice treatment algorithm but also integrating multiple auditory artificial models of electronic cochlea voice treatment algorithm for conveniently confirming parameter standard and evaluation rule to compare and estimate voice treatment result of different algorithm.
Description
Technical field
The present invention relates to the software platform that emulation is carried out in a kind of electric sense of hearing that cochlear implant implantation patient is produced, especially can integrate the sense of hearing realistic model of multiple cochlear implant speech processing algorithm and the emulation platform of evaluation and test algorithm effect.
Background technology
Cochlear implant is a kind of electronic installation that substitutes normal person's inner ear function, is used to help anacusia patient to recover the sense of hearing.The cochlear implant utilization is implanted the electrode of inner ear and is walked around the inner ear damaged part, and it is excited directly to stimulate remaining auditory nerve to produce with weak current, and this excitement is passed to brain and is interpreted as sound along Auditory Pathway, thereby makes the patient obtain the sense of hearing.Because what the patient heard is not the sound of real sound field, but by the sense of hearing of electric stimulus inducing, so be called for short the electric sense of hearing.Whether really the quality of electricity auditory effect feels similar with the normal person hears sound, depends on the quality of the speech coding algorithm that moves in the external speech coder of cochlear implant.Yet, can not rely on repeatedly to implant on one's body the patient repeatedly and verify and selection algorithm.Therefore, very need to adopt the method for setting up sense of hearing realistic model to debug, estimate, verify a kind of research platform of various algorithms.
Voice signal handling principle according to position-tone principle, frequency-tone principle, sensation of loudness principle scheduling theory and the cochlear implant of cochlea, cochlea is under the effect that weak current stimulates, near the electrode neuron is activated simultaneously, is equivalent to the discontinuous frequency-domain analysis of voice signal.Stimulate the diverse location of cochlea to cause the tone sense that brain is different, different stimulation amplitudes causes the different sensations of loudness, brain to these information processing after, produce the voice sensation, Here it is electric sense of hearing composition principle.Sense of hearing realistic model is done with its corresponding simplified the principle formula that the electric sense of hearing is synthetic according to the treatment scheme of different phonetic Processing Algorithm, thereby obtains the sense of hearing realistic model based on algorithms of different.
At present, in the research of existing cochlear implant speech processing algorithm, what propose and use mainly is the sense of hearing realistic model of building at single cochlear implant speech processing algorithm.But, what the different cochlear implant products that use clinically adopted respectively at present is different speech processing algorithms, only depend on and set up single sense of hearing realistic model and come very difficulty of speech processing algorithm quality that the comparison different product adopted, essentially no legally constituted authority one parameter and standard and interpretational criteria, and existing sense of hearing realistic model is not contained all cochlear implant speech processing algorithms, and the simulation parameter that is comprised does not have to characterize the actual features of cochlear implant product comprehensively yet.These defectives all are unfavorable for the research of cochlear implant speech processing algorithm and the exploitation of improvement and new algorithm.
Summary of the invention
Can't the lateral comparison algorithm effect in order to overcome the existing sense of hearing realistic model of setting up at single algorithm, and the limited deficiency of simulation parameter that comprises of sense of hearing realistic model, the invention provides a kind of cochlear implant sense of hearing emulation platform, this sense of hearing emulation platform can not only dope the electric auditory effect of the electronic cochlea implant persons that uses different cochlear implant speech processing algorithms, and integrated the sense of hearing realistic model of multiple cochlear implant speech processing algorithm, can determine parameter and standard and interpretational criteria easily, thereby the voice processing effect of lateral comparison and the algorithm that adopted of the different cochlear implant hardware product of evaluation and test, and algorithm research and development platform are easily provided for the researchist.
Cochlear implant sense of hearing emulation platform of the present invention comprises:
One sense of hearing realistic model routine library, it has integrated in the existing cochlear implant product employed multiple speech processing algorithm and based on the pairing sense of hearing realistic model of the speech processing algorithm of feature of Chinese language program;
One phonetic material database, some raw tone materials are arranged in its storage and corresponding to several emulation phonetic material of each raw tone material, described several emulation phonetic material corresponding to some raw tone materials are to be obtained through the sense of hearing realistic model routine processes of corresponding speech processing algorithm respectively by this raw tone material;
One sense of hearing l-G simulation test database, it comprises sense of hearing l-G simulation test test item bank, sense of hearing Simulation results storehouse and sense of hearing l-G simulation test experimenter information bank; And,
One sense of hearing simulation Software Platform, this simulation Software Platform comprise,
One sense of hearing l-G simulation test module, examination question in the described sense of hearing l-G simulation test of this module invokes database, normal good hearing crowd by tested person pair discerns, differentiates with the corresponding emulation phonetic material of handling through sense of hearing realistic model of this examination question, predict electronic cochlea implantation person's electric auditory effect, and the recognition correct rate of statistical results is evaluated and tested the effect of corresponding speech processing algorithm; With,
One speech processing algorithm comparison module, the described sense of hearing realistic model of this module invokes routine library, phonetic material database and sense of hearing l-G simulation test database carry out between the different phonetic Processing Algorithm of cochlear implant product relatively and between algorithm different parameters of the same race relatively, realize the speech processing algorithm lateral comparison.
Sense of hearing l-G simulation test examination question in the described sense of hearing l-G simulation test database, comprise two types of individual event selection examination question and open-ended questions, contents of test question comprises the examination question at the examination question of the emulation phonetic material of individual voice Processing Algorithm and different phonetic Processing Algorithm emulation phonetic material aliasing, and the examination question of the emulation phonetic material of the individual voice Processing Algorithm of different signal to noise ratio (S/N ratio)s and the examination question of different phonetic Processing Algorithm emulation phonetic material aliasing under the white noise background.
Sense of hearing simulating experimental based on above-mentioned cochlear implant sense of hearing emulation platform may further comprise the steps:
1., experimenter's initial landing cochlear implant sense of hearing emulation platform of normal good hearing, register or upgrade described experimenter's information;
2., select the examination question type of sense of hearing l-G simulation test, beginning sense of hearing l-G simulation test, sense of hearing simulation Software Platform accesses the examination question that described experimenter selects from sense of hearing l-G simulation test test item bank, play the emulation phonetic material of per pass examination question correspondence to described experimenter, the experimenter carries out identification according to the emulation phonetic material of hearing, selects or fill in corresponding answer simultaneously;
3., after examination question all answered, described sense of hearing simulation Software Platform is marked to test findings, adds up described experimenter's recognition correct rate, upgrades sense of hearing Simulation results storehouse simultaneously.
Based on the speech processing algorithm comparative approach of above-mentioned cochlear implant sense of hearing emulation platform, comprise between the different phonetic Processing Algorithm and relatively and between speech processing algorithm different parameters of the same race comparing, relatively may further comprise the steps between described different phonetic Processing Algorithm:
A, selection speech processing algorithm to be compared;
B, prediction algorithms of different electricity auditory effect relatively carry out the contrast of algorithms of different emulation voice signal time domain and frequency-domain waveform and carry out raw tone and the comparison of tin distinguishing of algorithms of different emulation sound effect;
C, speech processing algorithm to be compared is carried out Simulation results relatively, l-G simulation test recognition correct rate and with the clinical statistics data relatively under the contrast algorithms of different with the test findings of algorithms of different;
D, speech processing algorithm to be compared is carried out the electro photoluminescence coding strategy relatively, the electro photoluminescence coding strategy under the algorithms of different is compared;
Relatively may further comprise the steps between described speech processing algorithm different parameters of the same race:
1), selects parameter;
2), carry out under the different parameters the electric auditory effect of prediction relatively, carry out the contrast of different parameters emulation voice signal time domain and frequency-domain waveform and carry out raw tone and the comparison of tin distinguishing of different parameters emulation sound effect;
3), select different parameters to carry out Simulation results relatively, l-G simulation test recognition correct rate and with the clinical statistics data relatively under the contrast different parameters with the test findings of different parameters.
The present invention is according to cochlear implant voice signal Processing Algorithm and sense of hearing emulation technology, utilize the VC++6.0 platform in conjunction with Matlab6.5 program and Microsoft Access 2003 database interfaces, by integrating sense of hearing realistic model routine library, phonetic material database and sense of hearing l-G simulation test database, realized cochlear implant sense of hearing emulation platform.Its main effect has:
A) provide the sense of hearing realistic model routine library of integrating multiple cochlear implant speech processing algorithm, can dope the implantation person's who uses different cochlear implant speech processing algorithms electric auditory effect easily.
B) can determine parameter and standard and interpretational criteria easily, stride the voice processing effect that the cochlear implant hardware platform comes comparison and evaluation and test different phonetic Processing Algorithm, for the researchist provides algorithm research and development platform easily.
C) provide and expanded the multi parameters control of sense of hearing realistic model algorithm, make sense of hearing realistic model can characterize the characteristic of cochlear implant product more all sidedly, thereby can study the cochlear implant speech processing algorithm more fully and effectively.
Description of drawings
Fig. 1 is this cochlear implant sense of hearing emulation platform one-piece construction block diagram;
Fig. 2 is the allomeric function block diagram of its sense of hearing simulation Software Platform;
Fig. 3 is that it carries out sense of hearing l-G simulation test and the result adds up schematic flow sheet;
Fig. 4 is its speech processing algorithm lateral comparison schematic flow sheet.
Embodiment
Below in conjunction with description of drawings the specific embodiment of the present invention.
Fig. 1 is the one-piece construction block diagram of this cochlear implant sense of hearing emulation platform.This cochlear implant sense of hearing emulation platform comprises following four parts: sense of hearing simulation Software Platform 1, sense of hearing realistic model routine library 4, phonetic material database 2 and sense of hearing l-G simulation test database 3.
Sense of hearing simulation Software Platform 1 is developed based on VC++6.0, be the basis and the user interactions environment of whole cochlear implant sense of hearing emulation platform, sense of hearing simulation Software Platform 1 is implemented in calling phonetic material database 2 in sense of hearing l-G simulation test module 11, speech processing algorithm comparison module 12, phonetic material database maintenance module 13, the sense of hearing l-G simulation test database maintenance module 14 by opening and read original phonetic material and emulation phonetic material file in the phonetic material database 2; By using the ADO database access technology of VC++6.0, be implemented in the User Information Database maintenance module 16 of sense of hearing l-G simulation test database 4 interface with User Information Database; To be packaged into com component based on the sense of hearing realistic model routine library 4 of Matlab6.5 exploitation and in VC++6.0, call the interface of realization and sense of hearing realistic model routine library 4 by the COM Builder technology of using Matlab6.5.Wherein, the emulation speech data in the phonetic material database 2 is obtained by the corresponding primary voice data of sense of hearing realistic model routine library 4 processing.
Sense of hearing realistic model routine library 4 has been integrated in the existing cochlear implant product employed multiple speech processing algorithm and based on the pairing sense of hearing realistic model of the speech processing algorithm of feature of Chinese language program.Sense of hearing realistic model program realizes under the Matlab6.5 programmed environment, principle according to the different phonetic Processing Algorithm, the information of from the raw tone material, extracting and transmitting to each cochlear implant implant electrode, the pitch frequency that comprises the raw tone material, the amplitude information of electro photoluminescence, stimulate pulsewidth, stimulating current amplitude and stimulation pulsewidth are to producing the influence of tone sense intensity, stimulation amplitude to the influence of tone and brain to tone sense of signal that this implant electrode transmits etc., information characteristics on each implant electrode is added altogether according to the phonetic synthesis rule, thereby extracted the main information of raw tone material, modulation synthesis of artificial voice.The input parameter of sense of hearing realistic model program is according to the principle of different phonetic Processing Algorithm, comprise accordingly: the branch band filter parameter of corresponding cochlear implant implant electrode number and position, extract the method for information characteristics in the raw tone material, the method for synthesis of artificial voice etc.With (the Continuous Interleaved Sampling that alternately takes a sample continuously, abbreviation CIS) sense of hearing realistic model program of algorithm is an example, number and position according to implant electrode, the raw tone material is carried out the frequency division tape handling and extracts the electro photoluminescence amplitude information of the amplitude information of frequency band as implant electrode, the tone sense of the brain that produces characterizes with the sinusoidal signal of one group of different frequency, the sinusoidal signal of the corresponding characteristic frequency in the position of implant electrode, the amplitude of the corresponding sinusoidal signal of the amplitude of stimulating current on the electrode, the variation of stimulation rates is corresponding to the change of sinusoidal frequency, and the sinusoidal signal that amplitude modulation(PAM) is crossed adds the emulation voice that must arrive the analog electrical sense of hearing altogether.
Phonetic material database 2 storage has some raw tone materials and corresponding to the emulation phonetic material of several analog electrical sense of hearings of each raw tone material, described raw tone material is the extraneous real speech signal of gathering, and described several emulation phonetic material corresponding to some raw tone materials are the synthesis of artificial voice that obtained through the sense of hearing realistic model routine processes of corresponding speech processing algorithm respectively by this raw tone material.Described raw tone material comprises phoneme, word and sentence material according to the mandarin sounding based on mandarin.
Sense of hearing l-G simulation test database 3 comprises sense of hearing l-G simulation test test item bank, sense of hearing Simulation results storehouse and sense of hearing l-G simulation test experimenter information bank based on Microsoft Access 2003 exploitations.Wherein, sense of hearing l-G simulation test examination question comprises that individual event selects two types of examination question and open-ended questions, contents of test question comprises the examination question that mixes at the examination question of individual voice Processing Algorithm and various speech processing algorithm, and the examination question of the phonetic material of different signal to noise ratio (S/N ratio)s and the examination question of different phonetic material aliasing under the white noise background.
Fig. 2 is the allomeric function block diagram of its sense of hearing simulation Software Platform 1.Described sense of hearing simulation Software Platform 1 is based on the VC++6.0 environment exploitation, realized the test platform of sense of hearing l-G simulation test by calling phonetic material database 2 and sense of hearing l-G simulation test database 3, and utilized the ADO database access technology of VC++6.0, COMBuilder technology and the VC++6.0 of Matlab6.5 to call the com component technology respectively, integrated sense of hearing l-G simulation test database 3 and sense of hearing realistic model routine library 4.As shown in Figure 2, sense of hearing simulation Software Platform 1 comprises following six modules: sense of hearing l-G simulation test module 11, speech processing algorithm comparison module 12, phonetic material database maintenance module 13, sense of hearing l-G simulation test database maintenance module 14, sense of hearing realistic model library maintenance module 15 and User Information Database maintenance module 16.
Described six modules are respectively cochlear implant voice signal Processing Algorithm research staff following six kinds of major functions are provided:
One, sense of hearing l-G simulation test module 11, carry out normal good hearing crowd's sense of hearing l-G simulation test, call the examination question in the described sense of hearing l-G simulation test database 3, the normal good hearing crowd of tested person couple discerns, differentiates with the corresponding emulation phonetic material of handling through sense of hearing realistic model that comprises the part voice messaging of examination question, predict electronic cochlea implantation person's electric auditory effect, and the recognition correct rate of statistical results is evaluated and tested the effect of corresponding speech processing algorithm.
Two, speech processing algorithm comparison module 12 carries out the speech processing algorithm lateral comparison, call between the different phonetic Processing Algorithm that described sense of hearing realistic model routine library 4, phonetic material database 2 and sense of hearing l-G simulation test database 3 carry out the cochlear implant product relatively and between algorithm different parameters of the same race relatively.To the information of transmitting on each cochlear implant implant electrode, extract its pitch frequency, electro photoluminescence amplitude function, stimulate pulsewidth, stimulating current amplitude and stimulate pulsewidth to the influence that produces tone sense intensity, stimulation amplitude to the influence of tone and brain tone sense to signal that this implant electrode transmits.
Three, phonetic material database maintenance module 13 provides the maintenance update of raw tone material and the maintenance update function of the emulation phonetic material of correspondence.The maintenance update step of described emulation phonetic material comprises, select the raw tone material, the speech processing algorithm that comprises in the selection sense of hearing realistic model routine library 4, call the pairing sense of hearing realistic model of speech processing algorithm program selected in the sense of hearing realistic model routine library 4, handle selected raw tone material, synthetic corresponding emulation phonetic material is upgraded the phonetic material database.
Four, sense of hearing l-G simulation test database maintenance module 14, carry out the maintenance update of sense of hearing l-G simulation test material and sense of hearing l-G simulation test examination question.Sense of hearing l-G simulation test examination question comprises that individual event selects two types of examination question and open-ended questions, contents of test question comprises the examination question at the examination question of the emulation phonetic material of individual voice Processing Algorithm and different phonetic Processing Algorithm emulation phonetic material aliasing, and the examination question of the emulation phonetic material of the individual voice Processing Algorithm of different signal to noise ratio (S/N ratio)s and the examination question of different phonetic Processing Algorithm emulation phonetic material aliasing under the white noise background.Sense of hearing l-G simulation test material is corresponding to the type and the content of sense of hearing l-G simulation test examination question, and individual event selects the test material of examination question based on the phoneme material, comprises vowel, consonant and tone, and the test material of open-ended question is based on word and sentence; The test material of the examination question of the emulation phonetic material of individual voice Processing Algorithm is the emulation phonetic material that the sense of hearing realistic model routine processes of same speech processing algorithm obtains, the aliasing of the emulation phonetic material that the test material of the examination question of different phonetic Processing Algorithm emulation phonetic material aliasing then obtains for the sense of hearing realistic model routine processes of speech processing algorithm, the white noise background that adds different signal to noise ratios in test material then obtain the test material of examination question of emulation phonetic material of the individual voice Processing Algorithm of different signal to noise ratios under the white noise background.
Five, sense of hearing realistic model library maintenance module 15 provides the maintenance function of sense of hearing realistic model routine library.The input parameter of sense of hearing realistic model program is according to the principle of different phonetic Processing Algorithm, comprise accordingly: the branch band filter parameter of corresponding cochlear implant implant electrode number and position, extract the method for information characteristics in the raw tone material, the method for synthesis of artificial voice etc.In sense of hearing realistic model library maintenance module 15, provide input parameter that the function of upgrading and preserving output is set to sense of hearing realistic model program.
Six, User Information Database maintenance module 16 carries out the User Information Database maintenance, by using the ADO database access technology of VC++6.0, open User Information Database, renewal, interpolation or deletion sense of hearing l-G simulation test experimenter and emulation platform keeper's information etc.
Fig. 3 carries out sense of hearing l-G simulation test and the result adds up schematic flow sheet.Sense of hearing simulating experimental based on above-mentioned cochlear implant sense of hearing emulation platform may further comprise the steps:
1., experimenter's initial landing cochlear implant sense of hearing emulation platform of normal good hearing, register or upgrade described experimenter's information S31;
2., select the examination question type S32 of sense of hearing l-G simulation test, beginning sense of hearing l-G simulation test S34, sense of hearing simulation Software Platform accesses the examination question S33 that described experimenter selects from sense of hearing l-G simulation test test item bank, utilize the voice play function of VC++6.0 to play the emulation phonetic material of per pass examination question correspondence to described experimenter simultaneously, the experimenter carries out identification according to the emulation phonetic material of hearing, selects or fill in corresponding answer simultaneously;
3., after examination question all answered, described sense of hearing simulation Software Platform is marked to test findings, adds up described experimenter's recognition correct rate S35, upgrades sense of hearing Simulation results storehouse S36 simultaneously.
Fig. 4 is a speech processing algorithm lateral comparison schematic flow sheet.The invention provides the speech processing algorithm lateral comparison function of striding the cochlear implant hardware platform, it is by comprehensive comparison that sense of hearing simulation means has realized the treatment effect of speech processing algorithm, comprises between the different phonetic Processing Algorithm of cochlear implant product relatively between S1 and speech processing algorithm different parameters of the same race relatively S2.
Wherein, comparing S1 between described different phonetic Processing Algorithm may further comprise the steps:
A, selection speech processing algorithm S11 to be compared;
B, prediction algorithms of different electricity auditory effect be S12 relatively, carry out the contrast S13 of algorithms of different emulation voice signal time domain and frequency-domain waveform and carry out raw tone with algorithms of different emulation sound effect tin distinguish comparison S14;
C, speech processing algorithm to be compared is carried out relatively S15 of Simulation results, under the contrast algorithms of different l-G simulation test recognition correct rate S16 and with the test findings of algorithms of different with clinical statistics data S17 relatively;
D, speech processing algorithm to be compared is carried out relatively S18 of electro photoluminescence coding strategy, the electro photoluminescence coding strategy under the algorithms of different is compared S19;
Comparing S2 between described speech processing algorithm different parameters of the same race may further comprise the steps:
1), selects parameter S 21;
2), carry out under the different parameters relatively S22 of the electric auditory effect of prediction, carry out the contrast S23 of different parameters emulation voice signal time domain and frequency-domain waveform and carry out raw tone with different parameters emulation sound effect tin distinguish comparison S24;
3), select different parameters to carry out relatively S25 of Simulation results, under the contrast different parameters l-G simulation test recognition correct rate S26 and with the test findings of different parameters with clinical statistics data S27 relatively.
Claims (8)
1, a kind of cochlear implant sense of hearing emulation platform is characterized in that comprising:
One sense of hearing realistic model routine library (4), it has integrated in the existing cochlear implant product employed multiple speech processing algorithm and based on the pairing sense of hearing realistic model of the speech processing algorithm of feature of Chinese language program;
One phonetic material database (2), some raw tone materials are arranged in its storage and corresponding to several emulation phonetic material of each raw tone material, described several emulation phonetic material corresponding to some raw tone materials are to be obtained through the sense of hearing realistic model routine processes of corresponding speech processing algorithm respectively by this raw tone material;
One sense of hearing l-G simulation test database (3), it comprises sense of hearing l-G simulation test test item bank, sense of hearing Simulation results storehouse and sense of hearing l-G simulation test experimenter information bank; And,
One sense of hearing simulation Software Platform (1), this simulation Software Platform (1) comprise,
One sense of hearing l-G simulation test module (11), examination question in the described sense of hearing l-G simulation test of this module invokes database (3), normal good hearing crowd by tested person pair discerns, differentiates with the corresponding emulation phonetic material of handling through sense of hearing realistic model of this examination question, predict electronic cochlea implantation person's electric auditory effect, and the recognition correct rate of statistical results is evaluated and tested the effect of corresponding speech processing algorithm; With,
One speech processing algorithm comparison module (12), the described sense of hearing realistic model of this module invokes routine library (4), phonetic material database (2) and sense of hearing l-G simulation test database (3) carry out between the different phonetic Processing Algorithm of cochlear implant product relatively and between algorithm different parameters of the same race relatively, realize the speech processing algorithm lateral comparison.
2, cochlear implant sense of hearing emulation platform according to claim 1, it is characterized in that: described sense of hearing simulation Software Platform (1) also comprises phonetic material database maintenance module (13), and this module (13) is used for the maintenance update of raw tone material and the maintenance update of the emulation phonetic material of correspondence; Wherein, the maintenance update of emulation phonetic material comprises, select the raw tone material, the speech processing algorithm that comprises in the selection sense of hearing realistic model routine library (4), call the pairing sense of hearing realistic model of speech processing algorithm program selected in the sense of hearing realistic model routine library (4), handle selected raw tone material, synthetic corresponding emulation phonetic material is upgraded the phonetic material database.
3, cochlear implant sense of hearing emulation platform according to claim 2, it is characterized in that: described sense of hearing simulation Software Platform (1) also comprises sense of hearing l-G simulation test database maintenance module (14), and this module (14) is used for the maintenance update to sense of hearing l-G simulation test material and examination question.
4, cochlear implant sense of hearing emulation platform according to claim 3, it is characterized in that: described sense of hearing simulation Software Platform (1) also comprises sense of hearing realistic model library maintenance module (15), and this module (15) is used for the parameter of sense of hearing realistic model program in time is provided with renewal.
5, cochlear implant sense of hearing emulation platform according to claim 4, it is characterized in that: described sense of hearing simulation Software Platform (1) also comprises User Information Database maintenance module (16), and this module (16) is used to upgrade sense of hearing l-G simulation test experimenter and emulation platform keeper's information.
6, according to the described cochlear implant sense of hearing of the arbitrary claim of claim 1-5 emulation platform, it is characterized in that: the sense of hearing l-G simulation test examination question in the described sense of hearing l-G simulation test database (3) comprises that individual event selects two types of examination question and open-ended questions, contents of test question comprises the examination question at the examination question of the emulation phonetic material of individual voice Processing Algorithm and different phonetic Processing Algorithm emulation phonetic material aliasing, and the examination question of the emulation phonetic material of the individual voice Processing Algorithm of different signal to noise ratio (S/N ratio)s and the examination question of different phonetic Processing Algorithm emulation phonetic material aliasing under the white noise background.
7, based on the sense of hearing simulating experimental of the described cochlear implant sense of hearing of claim 1 emulation platform, it is characterized in that may further comprise the steps:
1., experimenter's initial landing cochlear implant sense of hearing emulation platform of normal good hearing, register or upgrade described experimenter's information;
2., select the examination question type of sense of hearing l-G simulation test, beginning sense of hearing l-G simulation test, sense of hearing simulation Software Platform accesses the examination question that described experimenter selects from sense of hearing l-G simulation test test item bank, play the emulation phonetic material of per pass examination question correspondence to described experimenter, the experimenter carries out identification according to the emulation phonetic material of hearing, selects or fill in corresponding answer simultaneously;
3., after examination question all answered, described sense of hearing simulation Software Platform is marked to test findings, adds up described experimenter's recognition correct rate, upgrades sense of hearing Simulation results storehouse simultaneously.
8, based on the speech processing algorithm comparative approach of the described cochlear implant sense of hearing of claim 1 emulation platform, it is characterized in that: comprise comparing between the different phonetic Processing Algorithm and compare (S2) between (S1) and speech processing algorithm different parameters of the same race,
Comparing (S1) between described different phonetic Processing Algorithm may further comprise the steps:
A, selection speech processing algorithm (S11) to be compared;
B, prediction algorithms of different electricity auditory effect be (S12) relatively, carry out the contrast (S13) of algorithms of different emulation voice signal time domain and frequency-domain waveform and carry out raw tone with algorithms of different emulation sound effect tin distinguish comparison (S14);
C, speech processing algorithm to be compared is carried out Simulation results relatively (S15), under the contrast algorithms of different l-G simulation test recognition correct rate (S16) and with the test findings of algorithms of different with clinical statistics data relatively (S17);
D, speech processing algorithm to be compared is carried out electro photoluminescence coding strategy relatively (S18), the electro photoluminescence coding strategy under the algorithms of different is compared (S19);
Comparing (S2) between described speech processing algorithm different parameters of the same race may further comprise the steps:
1), selects parameter (S21);
2), carry out under the different parameters the electric auditory effect of prediction relatively (S22), carry out the contrast (S23) of different parameters emulation voice signal time domain and frequency-domain waveform and carry out raw tone with different parameters emulation sound effect tin distinguish comparison (S24);
3), select different parameters to carry out Simulation results relatively (S25), under the contrast different parameters l-G simulation test recognition correct rate (S26) and with the test findings of different parameters with clinical statistics data relatively (S27).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100605796A CN100399275C (en) | 2006-05-15 | 2006-05-15 | Electronic canails audition simulating platform, its simulating experimental method and algorithm comparation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100605796A CN100399275C (en) | 2006-05-15 | 2006-05-15 | Electronic canails audition simulating platform, its simulating experimental method and algorithm comparation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1896955A true CN1896955A (en) | 2007-01-17 |
| CN100399275C CN100399275C (en) | 2008-07-02 |
Family
ID=37609483
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006100605796A Expired - Fee Related CN100399275C (en) | 2006-05-15 | 2006-05-15 | Electronic canails audition simulating platform, its simulating experimental method and algorithm comparation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100399275C (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101926692A (en) * | 2010-07-09 | 2010-12-29 | 清华大学 | Speech signal processor for cochlear implant based on dual core processor |
| CN101953722A (en) * | 2010-09-10 | 2011-01-26 | 中国科学院声学研究所 | Cochlear implant and sound signal processing method thereof |
| CN102670331A (en) * | 2012-06-04 | 2012-09-19 | 清华大学 | Electronic cochlear in-vitro debugging platform based on local area network (LAN) |
| CN104783926B (en) * | 2014-01-20 | 2016-09-28 | 中国科学院声学研究所 | A kind of quickly multiple tracks synchronizes cochlear implant equipment performance assessment system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1996006586A1 (en) * | 1994-08-26 | 1996-03-07 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Cochlea implant system |
| CN2619307Y (en) * | 2003-04-30 | 2004-06-02 | 合肥工业大学 | Analogue simulation system of machine visible computer |
-
2006
- 2006-05-15 CN CNB2006100605796A patent/CN100399275C/en not_active Expired - Fee Related
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101926692A (en) * | 2010-07-09 | 2010-12-29 | 清华大学 | Speech signal processor for cochlear implant based on dual core processor |
| CN101953722A (en) * | 2010-09-10 | 2011-01-26 | 中国科学院声学研究所 | Cochlear implant and sound signal processing method thereof |
| CN101953722B (en) * | 2010-09-10 | 2012-11-21 | 中国科学院声学研究所 | Cochlear implant and sound signal processing method thereof |
| CN102670331A (en) * | 2012-06-04 | 2012-09-19 | 清华大学 | Electronic cochlear in-vitro debugging platform based on local area network (LAN) |
| CN102670331B (en) * | 2012-06-04 | 2014-12-31 | 清华大学 | Electronic cochlear in-vitro debugging platform based on local area network (LAN) |
| CN104783926B (en) * | 2014-01-20 | 2016-09-28 | 中国科学院声学研究所 | A kind of quickly multiple tracks synchronizes cochlear implant equipment performance assessment system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100399275C (en) | 2008-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Newman et al. | Closed-loop, multichannel experimentation using the open-source NeuroRighter electrophysiology platform | |
| CN113288183B (en) | Silent voice recognition method based on facial neck surface myoelectricity | |
| Norton et al. | “Bird song Metronomics”: isochronous organization of zebra finch song rhythm | |
| US9826929B2 (en) | Method and device for stuttering alleviation | |
| Munkong et al. | Auditory perception and cognition | |
| Fu et al. | Effects of auditory training on adult cochlear implant patients: a preliminary report | |
| CN109036569B (en) | Verification method of time domain fine structure novel coding of artificial cochlea tone language | |
| Toiviainen et al. | Timbre similarity: Convergence of neural, behavioral, and computational approaches | |
| Brochier et al. | From microphone to phoneme: an end-to-end computational neural model for predicting speech perception with cochlear implants | |
| CN1896955A (en) | Electronic canails audition simulating platform, its simulating experimental method and algorithm comparation | |
| CN105528611A (en) | Ache identification classifier training method and device | |
| CN103239237A (en) | Tinnitus diagnostic test device | |
| CN105930480A (en) | Method for generating tinnitus rehabilitation music and tinnitus rehabilitation system | |
| Morello et al. | Dysphonia and dysarthria in people with Parkinson's disease after subthalamic nucleus deep brain stimulation: effect of frequency modulation | |
| Plack et al. | Overview: The present and future of pitch | |
| Arneodo et al. | Neurally driven synthesis of learned, complex vocalizations | |
| US20040088164A1 (en) | Voice connection system between humans and animals | |
| Alvarez et al. | A computational model to simulate spectral modulation and speech perception experiments of cochlear implant users | |
| Gómez-Vilda et al. | Neuromorphic detection of speech dynamics | |
| Rao et al. | Sensory feedback independent pre-song vocalizations correlate with time to song initiation | |
| CN109411076A (en) | A kind of emotion with preference consoles tree generation method and its emotion interactive system | |
| CN114969379A (en) | Big data-based method and device for generating single disease health management knowledge map | |
| Zamaninezhad et al. | A physiologically-inspired model reproducing the speech intelligibility benefit in cochlear implant listeners with residual acoustic hearing | |
| Uttal | Cellular neurophysiology and integration: an interpretive introduction | |
| Kuczapski et al. | Modeling and Simulation of Hearing with Cochlear Implants: A Proposed Method for Better Auralization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080702 Termination date: 20170515 |