CN105167883A - Speech processing method and system for increasing Chinese tone recognition rate based on frequency shift processing - Google Patents
Speech processing method and system for increasing Chinese tone recognition rate based on frequency shift processing Download PDFInfo
- Publication number
- CN105167883A CN105167883A CN201510624049.9A CN201510624049A CN105167883A CN 105167883 A CN105167883 A CN 105167883A CN 201510624049 A CN201510624049 A CN 201510624049A CN 105167883 A CN105167883 A CN 105167883A
- Authority
- CN
- China
- Prior art keywords
- fundamental frequency
- fixed electrode
- electrode
- sequence
- moved
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 11
- 238000000034 method Methods 0.000 claims abstract description 25
- 230000004936 stimulating effect Effects 0.000 claims abstract description 21
- 230000000638 stimulation Effects 0.000 claims abstract description 8
- 241000237858 Gastropoda Species 0.000 claims description 47
- 238000000605 extraction Methods 0.000 claims description 26
- 210000003477 cochlea Anatomy 0.000 claims description 22
- 238000001514 detection method Methods 0.000 claims description 3
- 230000005236 sound signal Effects 0.000 abstract 2
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000008447 perception Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 150000001875 compounds Chemical class 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 230000000630 rising effect Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 239000007943 implant Substances 0.000 description 1
- 238000002513 implantation Methods 0.000 description 1
- 230000007794 irritation Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000033764 rhythmic process Effects 0.000 description 1
Landscapes
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
- Electrotherapy Devices (AREA)
Abstract
The invention discloses a speech processing method and system for increasing the Chinese tone recognition rate based on frequency shift processing. The method comprises the following steps that M usable electrodes and H movable electrodes are determined, and therefore M-H electrodes exist in a fixed electrode sequence; changes of the fundamental frequency are detected according to the reference fundamental frequency and are expressed in percentage; when H is 2, the fixed electrode sequence is shifted if the change of the fundamental frequency exceeds 20% of the reference fundamental frequency; when H is 4, the fixed electrode sequence is shifted if the change of the fundamental frequency exceeds 15% of the reference fundamental frequency; when H is 6, the fixed electrode sequence is shifted if the change of the fundamental frequency exceeds 10% of the reference fundamental frequency; after the fixed electrode sequence is determined, N channels with the maximum energy are selected to be subjected to stimulation. According to the speech processing method and system for increasing the Chinese tone recognition rate based on frequency shift processing, the integral position of stimulating electrodes is changed according to the changes of the fundamental frequency of input sound signals so that the information of frequency changes in the sound signals can be transmitted, and finally the result of increasing the Chinese tone recognition rate is achieved.
Description
Technical field
The invention belongs to medical instruments field, particularly a kind of speech processing method and system improving Chinese tone recognition rate based on shift frequency process.
Background technology
For English speech recognition, the fundamental frequency information of the sound wave in speech production and the contribution of periodical information more weak, but for the Tone recognition of Chinese, fundamental frequency information and periodical information are even more important.
The identification of Chinese language tone is mainly realized by the change of perception frequency domain fundamental frequency (F0).Some other acoustic characteristic, also can be different under not same tone, and these acoustic featuress also may be helpful to the identification of tone.Such as, simple or compound vowel of a Chinese syllable length: in general, the simple or compound vowel of a Chinese syllable length of the 3rd sound is the longest, and the falling tone is the shortest; Sound amplitude: the peak amplitude of the 3rd sound is lower, and the peak amplitude of the falling tone is the highest; Amplitude envelops and F0 profile dependency: a lot of amplitude of variation envelope of sound and the change profile of fundamental frequency have dependency to a certain degree.Deposit in case at frequency domain fundamental frequency (F0), these acoustic characteristics are very little on Tone recognition impact.Therefore, pitch contour is the identity authentication of speaker, and rhythm identification and Tone recognition provide important information.
In artificial cochlea system, large quantifier elimination proves, although there is larger individual variation, tone (frequency) perception of implantation person is sharp from snail with implant electrode is arrange from low to high at the bottom of snail, conforms to the place coding theory of normal hearing system medium pitch perception.Tradition coding strategy is when carrying out electricity irritation to cochlea, and the range of choice of electrode is fixing, and cause for comparatively property is poor, can not transmit the information of acoustical signal pitch variation, power consumption is high.
Summary of the invention
In view of this, the object of the present invention is to provide a kind of the speech processing method and the system that improve Chinese tone recognition rate based on shift frequency process, according to the change of input acoustical signal frequency, suitably change the integral position of stimulating electrode, thus reach the effect improving Chinese tone recognition.The present invention is mainly for artificial cochlea, therefore content below is all treated to basis with artificial cochlea's speech.
For achieving the above object, the invention provides a kind of artificial cochlea's speech processing method improving Chinese tone recognition rate based on shift frequency process, comprising the following steps:
Determine M avaivable electrode, H float electrode, in fixed electrode sequence, have M-H electrode;
Carry out fundamental frequency extraction to the acoustical signal of input language processing device for artificial cochlea, obtain stable reference fundamental frequency, acquiescence fixed electrode sequence is in the centre position between snail hearth electrode to cupula cochleae electrode, is the default location of fixed electrode sequence herein;
Detect pitch variation, be expressed as a percentage;
When H is 2, pitch variation higher than with reference to fundamental frequency 20% time, fixed electrode sequence is moved one to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 20% time, fixed electrode sequence is moved one to cupula cochleae electrode direction;
When H is 4, pitch variation higher than with reference to fundamental frequency 15% time, fixed electrode sequence is moved one to snail hearth electrode direction; Higher than with reference to fundamental frequency 30% time, fixed electrode sequence is moved two to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 15% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; Lower than with reference to fundamental frequency 30% time, fixed electrode sequence is moved two to cupula cochleae electrode direction;
When H is 6, pitch variation higher than with reference to fundamental frequency 10% time, fixed electrode sequence is moved one to snail hearth electrode direction; Higher than with reference to fundamental frequency 20% time, fixed electrode sequence is moved two to snail hearth electrode direction; Higher than with reference to fundamental frequency 30% time, fixed electrode sequence is moved three to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 10% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; Pitch variation lower than with reference to fundamental frequency 20% time, fixed electrode sequence is moved two to cupula cochleae electrode direction; Pitch variation lower than with reference to fundamental frequency 30% time, fixed electrode sequence is moved three to cupula cochleae electrode direction;
After determining fixed electrode sequence, the N number of passage choosing energy maximum stimulates.
Preferably, described in carry out fundamental frequency extraction, obtaining stable reference fundamental frequency, for extracting the stable fundamental frequency of 3 frames continuously, being defined as with reference to fundamental frequency.
Preferably, after described detection pitch variation, corresponding movement is immediately made with fundamental frequency.
Preferably, when H is 2, after stimulating after 200ms, fixed electrode sequence is switched back default location.
Preferably, when H is 4, after stimulating after 200ms, fixed electrode sequence is made corresponding movement immediately with fundamental frequency.
Preferably, when H is 6, after stimulating after 200ms, fixed electrode sequence is made corresponding movement immediately with fundamental frequency.
Based on above-mentioned purpose, present invention also offers a kind of artificial cochlea's speech processing system improving Chinese tone recognition rate based on shift frequency process, comprising:
Electrode chooses unit, and for determining M avaivable electrode, H float electrode, has M-H electrode in fixed electrode sequence;
Fundamental frequency extraction unit, choose unit with described electrode to be connected, for carrying out fundamental frequency extraction to the acoustical signal of input language processing device for artificial cochlea, obtain stable reference fundamental frequency, acquiescence fixed electrode sequence is in the centre position between snail hearth electrode to cupula cochleae electrode, is the default location of fixed electrode sequence herein; Then detect pitch variation, be expressed as a percentage;
Shift frequency decision package, is connected with described fundamental frequency extraction unit, for different according to float electrode number H, carries out shift frequency process, when H is 2, pitch variation higher than with reference to fundamental frequency 20% time, fixed electrode sequence is moved one to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 20% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; When H is 4, pitch variation higher than with reference to fundamental frequency 15% time, fixed electrode sequence is moved one to snail hearth electrode direction; Higher than with reference to fundamental frequency 30% time, fixed electrode sequence is moved two to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 15% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; Lower than with reference to fundamental frequency 30% time, fixed electrode sequence is moved two to cupula cochleae electrode direction; When H is 6, pitch variation higher than with reference to fundamental frequency 10% time, fixed electrode sequence is moved one to snail hearth electrode direction; Higher than with reference to fundamental frequency 20% time, fixed electrode sequence is moved two to snail hearth electrode direction; Higher than with reference to fundamental frequency 30% time, fixed electrode sequence is moved three to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 10% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; Pitch variation lower than with reference to fundamental frequency 20% time, fixed electrode sequence is moved two to cupula cochleae electrode direction; Pitch variation lower than with reference to fundamental frequency 30% time, fixed electrode sequence is moved three to cupula cochleae electrode direction;
Passage stimulating unit, is connected with described shift frequency decision package, and for after determining fixed electrode sequence, the N number of passage choosing energy maximum stimulates.
Preferably, described fundamental frequency extraction unit, for extracting the stable fundamental frequency of 3 frames continuously, is defined as with reference to fundamental frequency.
Preferably, described fundamental frequency extraction unit, after detecting pitch variation, makes corresponding movement immediately with fundamental frequency.
Preferably, described shift frequency decision package, when H is 2, after stimulating after 200ms, switches back default location by fixed electrode sequence; When H is 4, after stimulating after 200ms, fixed electrode sequence is made corresponding movement immediately with fundamental frequency; When H is 6, after stimulating after 200ms, fixed electrode sequence is made corresponding movement immediately with fundamental frequency.
Beneficial effect of the present invention is: generally, and speech base frequency range is 100-350Hz, and the pitch variation scope of the Chinese four tones of standard Chinese pronunciation has the change scope of nearly 10dB because of men and women's sound difference.Wherein, the four tones of standard Chinese pronunciation pitch variation scope of female voice is respectively: first sound 5.3%; The rising tone 40.0%; 3rd sound 31.5%; The falling tone 56.6%; Male voice four tones of standard Chinese pronunciation pitch variation scope is respectively: first sound 7%; The rising tone 39.4%; 3rd sound 27.4%; The falling tone 55.5%, so the pitch variation scope of the Chinese four tones of standard Chinese pronunciation is comparatively large, if all electrode of unified use sends stimulation, does not have specific aim, bad to the effect of Tone recognition.To the dependency of the optimization of electrode stimulating only by increase rate envelope and fundamental frequency in prior art, stimulate with channel adaptive dynamic peak value, the present invention is by introducing fixed electrode and float electrode, only use fixed electrode sequence, according to the change of acoustical signal frequency, suitable change fixed electrode sequence location, stimulates different frequency perceived position, thus improves the Chinese tone recognition rate of language processing device for artificial cochlea; Meanwhile, owing to only using fixed electrode, power consumption is effectively reduced; Can choose with continuous alternate samples or peak value or tunnel strategy conbined usage after shift frequency decision package, there is adaptability and practicality widely.
Accompanying drawing explanation
In order to make object of the present invention, technical scheme and beneficial effect clearly, the invention provides following accompanying drawing and being described:
Fig. 1 is the flow chart of steps of artificial cochlea's speech processing method of the Chinese tone recognition rate that improves based on shift frequency process of the embodiment of the present invention;
Fig. 2 is the structural representation of artificial cochlea's speech processing system of the Chinese tone recognition rate that improves based on shift frequency process of the embodiment of the present invention;
Fig. 3 is the shift frequency stimulation schematic diagram improving float electrode H=2 in artificial cochlea's speech processing system of Chinese tone recognition rate based on shift frequency process of the embodiment of the present invention;
Fig. 4 is the shift frequency stimulation schematic diagram improving float electrode H=4 in artificial cochlea's speech processing system of Chinese tone recognition rate based on shift frequency process of the embodiment of the present invention;
Fig. 5 is the shift frequency stimulation schematic diagram improving float electrode H=6 in artificial cochlea's speech processing system of Chinese tone recognition rate based on shift frequency process of the embodiment of the present invention.
Detailed description of the invention
Below in conjunction with accompanying drawing, the preferred embodiments of the present invention are described in detail.
See Fig. 1, be depicted as the flow chart of steps of artificial cochlea's speech processing method of the Chinese tone recognition rate that improves based on shift frequency process of the embodiment of the present invention, comprise the following steps:
S101, determines M avaivable electrode, and H float electrode has M-H electrode in fixed electrode sequence;
S102, carries out fundamental frequency extraction to the acoustical signal of input language processing device for artificial cochlea, obtains stable reference fundamental frequency, and acquiescence fixed electrode sequence is in the centre position between snail hearth electrode to cupula cochleae electrode, is the default location of fixed electrode sequence herein;
S103, detects pitch variation, is expressed as a percentage;
S104, when H is 2, pitch variation higher than with reference to fundamental frequency 20% time, fixed electrode sequence is moved one to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 20% time, fixed electrode sequence is moved one to cupula cochleae electrode direction;
S105, when H is 4, pitch variation higher than with reference to fundamental frequency 15% time, fixed electrode sequence is moved one to snail hearth electrode direction; Higher than with reference to fundamental frequency 30% time, fixed electrode sequence is moved two to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 15% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; Lower than with reference to fundamental frequency 30% time, fixed electrode sequence is moved two to cupula cochleae electrode direction;
S106, when H is 6, pitch variation higher than with reference to fundamental frequency 10% time, fixed electrode sequence is moved one to snail hearth electrode direction; Higher than with reference to fundamental frequency 20% time, fixed electrode sequence is moved two to snail hearth electrode direction; Higher than with reference to fundamental frequency 30% time, fixed electrode sequence is moved three to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 10% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; Pitch variation lower than with reference to fundamental frequency 20% time, fixed electrode sequence is moved two to cupula cochleae electrode direction; Pitch variation lower than with reference to fundamental frequency 30% time, fixed electrode sequence is moved three to cupula cochleae electrode direction;
S107, after determining fixed electrode sequence, the N number of passage choosing energy maximum stimulates.
Further, described in S102, carrying out fundamental frequency extraction, obtain stable reference fundamental frequency, for extracting the stable fundamental frequency of 3 frames continuously, being defined as with reference to fundamental frequency.
Further, after described S103 step, corresponding movement is immediately made with fundamental frequency.
Further, when H is 2, after stimulating after 200ms, fixed electrode sequence is switched back default location.
Further, when H is 4, after stimulating after 200ms, fixed electrode sequence is made corresponding movement immediately with fundamental frequency.
Further, when H is 6, after stimulating after 200ms, fixed electrode sequence is made corresponding movement immediately with fundamental frequency.
With above-mentioned based on shift frequency process improve artificial cochlea's speech processing method of Chinese tone recognition rate corresponding be, further embodiment of this invention provides the structural representation of the artificial cochlea's speech processing system improving Chinese tone recognition rate based on shift frequency process, comprising:
Electrode chooses unit 101, and for determining M avaivable electrode, H float electrode, has M-H electrode in fixed electrode sequence;
Fundamental frequency extraction unit 102, choose unit 101 with electrode to be connected, for carrying out fundamental frequency extraction to the acoustical signal of input language processing device for artificial cochlea, obtain stable reference fundamental frequency, acquiescence fixed electrode sequence is in the centre position between snail hearth electrode to cupula cochleae electrode, is the default location of fixed electrode sequence herein; Then detect pitch variation, be expressed as a percentage;
Shift frequency decision package 103, is connected with fundamental frequency extraction unit 102, for different according to float electrode number H, carries out shift frequency process, when H is 2, pitch variation higher than with reference to fundamental frequency 20% time, fixed electrode sequence is moved one to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 20% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; When H is 4, pitch variation higher than with reference to fundamental frequency 15% time, fixed electrode sequence is moved one to snail hearth electrode direction; Higher than with reference to fundamental frequency 30% time, fixed electrode sequence is moved two to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 15% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; Lower than with reference to fundamental frequency 30% time, fixed electrode sequence is moved two to cupula cochleae electrode direction; When H is 6, pitch variation higher than with reference to fundamental frequency 10% time, fixed electrode sequence is moved one to snail hearth electrode direction; Higher than with reference to fundamental frequency 20% time, fixed electrode sequence is moved two to snail hearth electrode direction; Higher than with reference to fundamental frequency 30% time, fixed electrode sequence is moved three to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 10% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; Pitch variation lower than with reference to fundamental frequency 20% time, fixed electrode sequence is moved two to cupula cochleae electrode direction; Pitch variation lower than with reference to fundamental frequency 30% time, fixed electrode sequence is moved three to cupula cochleae electrode direction;
Passage stimulating unit 104, is connected with shift frequency decision package 103, and for after determining fixed electrode sequence, the N number of passage choosing energy maximum stimulates.
Further, in embody rule example, fundamental frequency extraction unit 102, for extracting the stable fundamental frequency of 3 frames continuously, is defined as with reference to fundamental frequency.
Further, in embody rule example, after fundamental frequency extraction unit 102 detects pitch variation, immediately make corresponding movement with fundamental frequency.
Further, in embody rule example, shift frequency decision package 103, when H is 2, after stimulating after 200ms, switches back default location by fixed electrode sequence; When H is 4, after stimulating after 200ms, fixed electrode sequence is made corresponding movement immediately with fundamental frequency; When H is 6, after stimulating after 200ms, fixed electrode sequence is made corresponding movement immediately with fundamental frequency.Switching back default location and immediately making with fundamental frequency in the process of corresponding movement, fundamental frequency extraction unit 102 no longer carries out fundamental frequency extraction.
The present invention is for the artificial cochlea of physical electrode more than 20.24 avaivable electrodes in specific embodiment, No. 1-24, electrode numbering, corresponding at the bottom of from cupula cochleae to snail, namely from low to high, 6 float electrodes, so fixed electrode is 18, and fixed electrode sequence is 4-21 electrode.Namely be divided into 18 passages, represent the first passage of lowest frequency with No. 4 electrode; The 24 passage of most high frequency is represented with No. 21 electrode.Each frame frequency division 18 passages, electrode number is designated as E1-E18.Under default mode, 4-21 electrode is corresponding E1-E18 respectively.
The selection of E1-E18 is continuous print.
The selection of E1-E18 changes with the change of input acoustical signal fundamental frequency.The limit is: corresponding 1 to the No. 18 electrode of E1-E18 or the corresponding 7-24 electrode of E1-E18.
In general, the float electrode number of (low frequency/high frequency) at the bottom of the cupula cochleae/snail of both sides wants equal symmetry.Float electrode number is not less than 2, is not more than 6.Fixed electrode number is not more than 22, is not less than 18.
In specific embodiment, be fundamental curve see two oscillograms of the first half in Fig. 3-5, figure, curve 1 is the size of fundamental frequency; Two coordinate diagram of the latter half represent fixed electrode sequence, and small circle 2 represents that this electrode is activated.
See Fig. 3, if float electrode is 2, namely 1 is right, is designated as 1 grade; After fundamental frequency extraction unit 102 obtains stable reference fundamental frequency, namely what pitch variation exceeded 20% starts shift frequency, if higher than with reference to 20% of fundamental frequency, fixed electrode sequence is moved to direction at the bottom of snail.
Fixed electrode sequence default location is the 2 to the No. 23 electrode, and float electrode is the 1st and No. 24 electrode.When detection pitch variation is original 120%, namely higher than during with reference to fundamental frequency 20%, small circle 2 displacement is the 3 to the No. 24 electrode; When detecting pitch variation and being original 80%, small circle 2 displacement is the 1 to the No. 22 electrode.
See Fig. 4, if 2 is right, be designated as 2 grades; After fundamental frequency extraction unit 102 obtains stable reference fundamental frequency, namely pitch variation starts shift frequency more than 15%, shift frequency direction is consistent with above-mentioned, move one higher than the 15% fixed electrode sequence with reference to fundamental frequency to the large electrode of numbering, move one lower than the 15% fixed electrode sequence with reference to fundamental frequency to the little electrode of numbering; Move one to the large electrode of numbering again higher than the 30% fixed electrode sequence with reference to fundamental frequency, move two compared with default location, move one lower than the 30% fixed electrode sequence with reference to fundamental frequency to the little electrode of numbering, move two compared with default location.Corresponding with the position of small circle 2 by the size variation of curve in Fig. 41, visible shift frequency effect is fine.
See Fig. 5, if 3 is right, be designated as 3 grades; After fundamental frequency extraction unit 102 obtains stable reference fundamental frequency, namely what pitch variation exceeded 10% starts shift frequency, shift frequency direction is consistent with above-mentioned, move one higher than the 10% fixed electrode sequence with reference to fundamental frequency to the large electrode of numbering, move one lower than the 10% fixed electrode sequence with reference to fundamental frequency to the little electrode of numbering; Move one to the large electrode of numbering again higher than the 20% fixed electrode sequence with reference to fundamental frequency, move two compared with default location, move one lower than the 20% fixed electrode sequence with reference to fundamental frequency to the little electrode of numbering, move two compared with default location; Move one to the large electrode of numbering again higher than the 30% fixed electrode sequence with reference to fundamental frequency, move three compared with default location, move one lower than the 30% fixed electrode sequence with reference to fundamental frequency to the little electrode of numbering, move three compared with default location.Corresponding with the position of small circle 2 by the size variation of curve in Fig. 51, visible shift frequency effect is fine.
What finally illustrate is, above preferred embodiment is only in order to illustrate technical scheme of the present invention and unrestricted, although by above preferred embodiment to invention has been detailed description, but those skilled in the art are to be understood that, various change can be made to it in the form and details, and not depart from claims of the present invention limited range.
Claims (10)
1. improve a speech processing method for Chinese tone recognition rate based on shift frequency process, it is characterized in that, comprise the following steps:
Determine M avaivable electrode, H float electrode, therefore have M-H electrode in fixed electrode sequence;
Carry out fundamental frequency extraction to the acoustical signal of input language processing device for artificial cochlea, obtain stable reference fundamental frequency, acquiescence fixed electrode sequence is in the centre position between snail hearth electrode to cupula cochleae electrode, is the default location of fixed electrode sequence herein;
Detect pitch variation, be expressed as a percentage;
When H is 2, pitch variation higher than with reference to fundamental frequency 20% time, fixed electrode sequence is moved one to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 20% time, fixed electrode sequence is moved one to cupula cochleae electrode direction;
When H is 4, pitch variation higher than with reference to fundamental frequency 15% time, fixed electrode sequence is moved one to snail hearth electrode direction; Higher than with reference to fundamental frequency 30% time, fixed electrode sequence is moved two to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 15% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; Lower than with reference to fundamental frequency 30% time, fixed electrode sequence is moved two to cupula cochleae electrode direction;
When H is 6, pitch variation higher than with reference to fundamental frequency 10% time, fixed electrode sequence is moved one to snail hearth electrode direction; Higher than with reference to fundamental frequency 20% time, fixed electrode sequence is moved two to snail hearth electrode direction; Higher than with reference to fundamental frequency 30% time, fixed electrode sequence is moved three to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 10% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; Pitch variation lower than with reference to fundamental frequency 20% time, fixed electrode sequence is moved two to cupula cochleae electrode direction; Pitch variation lower than with reference to fundamental frequency 30% time, fixed electrode sequence is moved three to cupula cochleae electrode direction;
After determining fixed electrode sequence, the N number of passage choosing energy maximum stimulates.
2. method according to claim 1, is characterized in that, described in carry out fundamental frequency extraction, obtaining stable reference fundamental frequency, for extracting the stable fundamental frequency of 3 frames continuously, being defined as with reference to fundamental frequency.
3. method according to claim 2, is characterized in that, after described detection pitch variation, electrode makes corresponding movement immediately with fundamental frequency.
4. method according to claim 1, is characterized in that, when H is 2, after stimulating after 200ms, fixed electrode sequence is switched back default location.
5. method according to claim 1, is characterized in that, when H is 4, after stimulating after 200ms, fixed electrode sequence is made corresponding movement immediately with fundamental frequency.
6. method according to claim 5, is characterized in that, when H is 6, after stimulating after 200ms, fixed electrode sequence is made corresponding movement immediately with fundamental frequency.
7. improve a speech processing system for Chinese tone recognition rate based on shift frequency process, it is characterized in that, comprising:
Electrode chooses unit, and for determining M avaivable electrode, H float electrode, has M-H electrode in fixed electrode sequence;
Fundamental frequency extraction unit, choose unit with described electrode to be connected, for carrying out fundamental frequency extraction to the acoustical signal of input language processing device for artificial cochlea, obtain stable reference fundamental frequency, acquiescence fixed electrode sequence is in the centre position between snail hearth electrode to cupula cochleae electrode, is the default location of fixed electrode sequence herein; Then detect pitch variation, be expressed as a percentage;
Shift frequency decision package, is connected with described fundamental frequency extraction unit, for different according to float electrode number H, carries out shift frequency process, when H is 2, pitch variation higher than with reference to fundamental frequency 20% time, fixed electrode sequence is moved one to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 20% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; When H is 4, pitch variation higher than with reference to fundamental frequency 15% time, fixed electrode sequence is moved one to snail hearth electrode direction; Higher than with reference to fundamental frequency 30% time, fixed electrode sequence is moved two to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 15% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; Lower than with reference to fundamental frequency 30% time, fixed electrode sequence is moved two to cupula cochleae electrode direction; When H is 6, pitch variation higher than with reference to fundamental frequency 10% time, fixed electrode sequence is moved one to snail hearth electrode direction; Higher than with reference to fundamental frequency 20% time, fixed electrode sequence is moved two to snail hearth electrode direction; Higher than with reference to fundamental frequency 30% time, fixed electrode sequence is moved three to snail hearth electrode direction; Pitch variation lower than with reference to fundamental frequency 10% time, fixed electrode sequence is moved one to cupula cochleae electrode direction; Pitch variation lower than with reference to fundamental frequency 20% time, fixed electrode sequence is moved two to cupula cochleae electrode direction; Pitch variation lower than with reference to fundamental frequency 30% time, fixed electrode sequence is moved three to cupula cochleae electrode direction;
Passage stimulating unit, is connected with described shift frequency decision package, and for after determining fixed electrode sequence, the N number of passage choosing energy maximum stimulates.
8. system according to claim 7, is characterized in that, described fundamental frequency extraction unit, after extracting the stable fundamental frequency of 3 frames continuously, is defined as by the fundamental frequency extracted with reference to fundamental frequency in the 3rd frame.
9. system according to claim 7, is characterized in that, described fundamental frequency extraction unit, after pitch variation being detected, immediately makes corresponding movement with fundamental frequency.
10. system according to claim 9, is characterized in that, described shift frequency decision package, when H is 2, after stimulation 200ms, fixed electrode sequence is switched back default location; When H is 4, after stimulation 200ms, fixed electrode sequence is made corresponding movement immediately with fundamental frequency; When H is 6, after stimulation 200ms, fixed electrode sequence is made corresponding movement immediately with fundamental frequency.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510624049.9A CN105167883B (en) | 2015-09-25 | 2015-09-25 | Speech processing method and system for increasing Chinese tone recognition rate based on frequency shift processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510624049.9A CN105167883B (en) | 2015-09-25 | 2015-09-25 | Speech processing method and system for increasing Chinese tone recognition rate based on frequency shift processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105167883A true CN105167883A (en) | 2015-12-23 |
| CN105167883B CN105167883B (en) | 2017-04-19 |
Family
ID=54890680
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510624049.9A Active CN105167883B (en) | 2015-09-25 | 2015-09-25 | Speech processing method and system for increasing Chinese tone recognition rate based on frequency shift processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105167883B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107507610A (en) * | 2017-09-28 | 2017-12-22 | 河南理工大学 | A kind of Chinese tone recognition method based on vowel fundamental frequency information |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040015356A1 (en) * | 2002-07-17 | 2004-01-22 | Matsushita Electric Industrial Co., Ltd. | Voice recognition apparatus |
| US20090112288A1 (en) * | 2007-10-25 | 2009-04-30 | Korea Institute Of Machinery & Materials | Frequency analyzer for mems based cochlear implant with self power supply |
| CN101803972A (en) * | 2010-01-11 | 2010-08-18 | 杭州诺尔康神经电子科技有限公司 | Cochlear implant system and automatic physical electrode selection method |
| US8019431B2 (en) * | 2008-06-02 | 2011-09-13 | University Of Washington | Enhanced signal processing for cochlear implants |
| US20120209351A1 (en) * | 2011-02-14 | 2012-08-16 | Med-El Elektromedizinische Geraete Gmbh | Enhancing Fine Time Structure Transmission for Hearing Implant System |
| CN104013492A (en) * | 2014-05-23 | 2014-09-03 | 浙江诺尔康神经电子科技股份有限公司 | Cochlear implant signal processing unit and method based on auditory perception prompts |
| CN104287872A (en) * | 2013-07-15 | 2015-01-21 | 奥迪康医疗有限公司 | A hearing assistance device comprising an implanted part for measuring and processing electrically evoked nerve responses |
| CN104616659A (en) * | 2015-02-09 | 2015-05-13 | 山东大学 | Method for influencing tone perception of reconstructed speech by phase and application in artificial cochlea |
-
2015
- 2015-09-25 CN CN201510624049.9A patent/CN105167883B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040015356A1 (en) * | 2002-07-17 | 2004-01-22 | Matsushita Electric Industrial Co., Ltd. | Voice recognition apparatus |
| US20090112288A1 (en) * | 2007-10-25 | 2009-04-30 | Korea Institute Of Machinery & Materials | Frequency analyzer for mems based cochlear implant with self power supply |
| US8019431B2 (en) * | 2008-06-02 | 2011-09-13 | University Of Washington | Enhanced signal processing for cochlear implants |
| CN101803972A (en) * | 2010-01-11 | 2010-08-18 | 杭州诺尔康神经电子科技有限公司 | Cochlear implant system and automatic physical electrode selection method |
| US20120209351A1 (en) * | 2011-02-14 | 2012-08-16 | Med-El Elektromedizinische Geraete Gmbh | Enhancing Fine Time Structure Transmission for Hearing Implant System |
| CN104287872A (en) * | 2013-07-15 | 2015-01-21 | 奥迪康医疗有限公司 | A hearing assistance device comprising an implanted part for measuring and processing electrically evoked nerve responses |
| CN104013492A (en) * | 2014-05-23 | 2014-09-03 | 浙江诺尔康神经电子科技股份有限公司 | Cochlear implant signal processing unit and method based on auditory perception prompts |
| CN104616659A (en) * | 2015-02-09 | 2015-05-13 | 山东大学 | Method for influencing tone perception of reconstructed speech by phase and application in artificial cochlea |
Non-Patent Citations (1)
| Title |
|---|
| 张宝琳 等: ""数字助听器中多通道响度补偿方法的研究"", 《信号处理》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107507610A (en) * | 2017-09-28 | 2017-12-22 | 河南理工大学 | A kind of Chinese tone recognition method based on vowel fundamental frequency information |
| CN107507610B (en) * | 2017-09-28 | 2020-08-07 | 河南理工大学 | Chinese tone recognition method based on vowel fundamental frequency information |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105167883B (en) | 2017-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Luo et al. | Enhancing Chinese tone recognition by manipulating amplitude envelope: Implications for cochlear implants | |
| CN101642399B (en) | Artificial cochlea speech processing method based on frequency modulation information and artificial cochlea speech processor | |
| US8185383B2 (en) | Methods and apparatus for adapting speech coders to improve cochlear implant performance | |
| CN105813688B (en) | Device for the transient state sound modification in hearing implant | |
| CN103140260A (en) | Cochlear implant stimulation with low frequency channel privilege | |
| CN101716108A (en) | System for processing electronic cochlear signal and coding electric stimulation and processing method thereof | |
| CN111768802B (en) | Artificial cochlea voice processing method and system | |
| CN102579159A (en) | Electrical cochlea speech processor and processing method with signal compression in wide dynamic range | |
| CN105769385A (en) | A cochlear implant and an operating method thereof | |
| CN103892939A (en) | Artificial cochlea speech processor and method for improving Chinese tone recognition rate | |
| CN107708797A (en) | Switch hearing implant coding strategy | |
| CN104307100B (en) | A kind of method and system improving artificial cochlea's pitch perception | |
| US7561709B2 (en) | Modulation depth enhancement for tone perception | |
| CN107708794A (en) | Selective stimulating with cochlear implant | |
| CN100502819C (en) | Artificial cochlea manufacture method suitable for Chinese voice coding strategy | |
| CN105167883A (en) | Speech processing method and system for increasing Chinese tone recognition rate based on frequency shift processing | |
| US11632634B2 (en) | Feature extraction in hearing prostheses | |
| CN101773429B (en) | Artificial electronic cochlea and method for processing speech with double stimulation rates | |
| Zheng et al. | A noise-robust signal processing strategy for cochlear implants using neural networks | |
| CN109200469B (en) | An optimized coding method and system for enhancing cochlear implant pitch perception | |
| CN109982744A (en) | Using reorganization fine structure processing based on the cochlear stimulation being concerned between ears | |
| CN104856784B (en) | A kind of electric auditory prosthesis signal processing method and its system | |
| CN101803972B (en) | Cochlear implant system and automatic physical electrode selection method | |
| US9100763B2 (en) | Determining stimulation signal strength in an auditory prosthesis | |
| CN105596120B (en) | Strengthen the cochlear electrode that music rhythm perceives and divide equally arrangement, device, system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |