CN110660396A - Language processing system and method based on MEMS - Google Patents
Language processing system and method based on MEMS Download PDFInfo
- Publication number
- CN110660396A CN110660396A CN201810607268.XA CN201810607268A CN110660396A CN 110660396 A CN110660396 A CN 110660396A CN 201810607268 A CN201810607268 A CN 201810607268A CN 110660396 A CN110660396 A CN 110660396A
- Authority
- CN
- China
- Prior art keywords
- mems
- target vector
- pulse
- unit
- processing system
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 18
- 238000001208 nuclear magnetic resonance pulse sequence Methods 0.000 claims abstract description 13
- 230000006870 function Effects 0.000 claims abstract description 4
- 238000006243 chemical reaction Methods 0.000 claims abstract description 3
- 230000007774 longterm Effects 0.000 claims description 9
- 238000007476 Maximum Likelihood Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 230000006698 induction Effects 0.000 claims description 3
- 230000035484 reaction time Effects 0.000 abstract description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/28—Constructional details of speech recognition systems
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
- G10L25/12—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters the extracted parameters being prediction coefficients
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The invention relates to a language processing system based on MEMS and a method thereof, which mainly comprises the following steps: the system comprises an MEMS (micro electro mechanical system) identification sensing unit, a target vector unit and a multi-pulse analysis unit, wherein the MEMS identification sensing unit is used for quantifying the multi-pulse analysis function; the target vector unit acts on a target vector, determines an initial gain level for the pulse sequence, completes single-gain multi-pulse analysis for multiple times, and provides a sequence which can represent the target vector most as an output signal for different gain levels each time; the output end of the MEMS identification sensing unit is connected with at least one pulse series multi-pulse analysis unit, the target vector unit can be used as a pulse series sequence, and the pulse series is composed of a group of single-gain pulses. The invention belongs to an ultra-short time analysis system, and greatly improves the reaction time and the reaction accuracy of a language processing system.
Description
Technical Field
The invention relates to a language processing system, in particular to a language processing system based on MEMS and a method thereof.
Background
At present, human beings have more and more closely related, and the language difference and other factors make the language processing system and method more and more important, so that the problems of the response time and the response accuracy of the language processing system are urgently needed to be solved.
Disclosure of Invention
In view of the above problems, an object of the present invention is to provide a MEMS-based language processing system and method thereof, so as to shorten the response time of the language processing system and improve the response accuracy.
In order to achieve the purpose, the invention adopts the technical scheme that:
a language processing system based on MEMS and its method includes:
MEMS discerns induction element: the output end of the MEMS identification sensing unit is connected with an analyzer, and the input end of the MEMS is used for inputting a speech signal to generate the ultrashort characteristic of the input speech signal;
target vector unit: the target vector unit generates a target vector by at least the input speech signal and the optional ultra-short time characteristic, and the multi-pulse analysis unit of the output line of the target vector unit is connected with the multi-pulse analysis unit of the output line of the target vector unit;
a multi-pulse analysis unit: the multi-pulse analysis unit generates a group of pulse sequences with equal amplitude, variable symbols and variable intervals, the multi-pulse analysis unit outputs a signal corresponding to one pulse sequence with equal amplitude, variable symbols and variable intervals corresponding to the multi-pulse analysis unit, and the sequence is a sequence which can judge that the sequence can represent the target vector most through judgment according to the maximum likelihood.
Further, a MEMS speech processing system and method thereof includes an ultra-short time analyzer for generating ultra-short time characteristics by performing linear prediction coefficients on an input speech signal.
Further, the initial pulse of each of the equal amplitude, variable sign, variable spacing burst sequences is located at the same position.
Further, the target vector unit comprises a global decision determiner comprising a perceptual weighting filter for filtering the constant amplitude, variable sign, variable spacing pulse sequences.
Further, the ultra-short time analyzer has functions and characteristics that combine MEMS recognition sensing and switching.
Furthermore, the input end of the MEMS identification sensing unit inputs a voice signal and can also generate long-term characteristics, the long-term characteristics at least comprise a tone value of the input voice signal and short-term characteristics of the MEMS identification sensing output end, and the target vector unit is used for generating a target vector at least by the input voice signal and optional short-term and long-term characteristics.
Further, the initial pulses of the pulse sequence are located at the same sample position.
The invention has the beneficial effects that:
the technical scheme of the invention is designed based on MEMS, belongs to an ultra-short time analysis system, greatly shortens the reaction time of a language processing system and improves the reaction accuracy.
Drawings
FIG. 1 is a schematic diagram of the MEMS-based language processing system and method of the present invention.
The specific implementation mode is as follows:
the present invention will be described in further detail below with reference to the accompanying drawings and technical solutions, and embodiments of the present invention will be described in detail by way of preferred examples, but the embodiments of the present invention are not limited thereto.
A language processing system based on MEMS and its method includes:
MEMS discerns induction element: the output end of the MEMS identification sensing unit is connected with an analyzer, and the input end of the MEMS is used for inputting a speech signal to generate the ultrashort characteristic of the input speech signal;
target vector unit: the target vector unit generates a target vector by at least the input speech signal and the optional ultra-short time characteristic, and the multi-pulse analysis unit of the output line of the target vector unit is connected with the multi-pulse analysis unit of the output line of the target vector unit;
a multi-pulse analysis unit: the multi-pulse analysis unit generates a group of pulse sequences with equal amplitude, variable symbols and variable intervals, the multi-pulse analysis unit outputs a signal corresponding to one pulse sequence with equal amplitude, variable symbols and variable intervals corresponding to the multi-pulse analysis unit, and the sequence is a sequence which can judge that the sequence can represent the target vector most through judgment according to the maximum likelihood.
Further, a MEMS speech processing system and method thereof includes an ultra-short time analyzer for generating ultra-short time characteristics by performing linear prediction coefficients on an input speech signal.
Further, the initial pulse of each of the equal amplitude, variable sign, variable spacing burst sequences is located at the same position.
Further, the target vector unit comprises a global decision determiner comprising a perceptual weighting filter for filtering the constant amplitude, variable sign, variable spacing pulse sequences.
Further, the ultra-short time analyzer has functions and characteristics that combine MEMS recognition sensing and switching.
Furthermore, the input end of the MEMS identification sensing unit inputs a voice signal and can also generate long-term characteristics, the long-term characteristics at least comprise a tone value of the input voice signal and short-term characteristics of the MEMS identification sensing output end, and the target vector unit is used for generating a target vector at least by the input voice signal and optional short-term and long-term characteristics.
Further, the initial pulses of the pulse sequence are located at the same sample position.
The above description is only a preferred embodiment of the present invention and is not limited to the present invention, it should be noted that, for those skilled in the art, many modifications and variations can be made without departing from the principle of the present invention, and these changes and variations should also be construed as the protection scope of the present invention.
Claims (7)
1. A language processing system based on MEMS and its method are characterized by comprising:
MEMS discerns induction element: the output end of the MEMS identification sensing unit is connected with an analyzer, and the input end of the MEMS is used for inputting a speech signal to generate the ultrashort characteristic of the input speech signal;
target vector unit: said target vector unit generating a target vector from at least said input speech signal and optionally said ultra short time characteristic, a multi-pulse analysis unit connected to an output line of said target vector unit and a multi-pulse analysis of an output line of said target vector unit;
a multi-pulse analysis unit: the multi-pulse analysis unit generates a group of pulse sequences with equal amplitude, variable symbols and variable intervals, and outputs a signal corresponding to the pulse sequences with equal amplitude, variable symbols and variable intervals with the multi-pulse analysis unit, wherein the sequences can be judged to represent the target vector most according to the maximum likelihood.
2. The language processing system of MEMS and the method thereof according to claim 1, wherein: an ultra-short time analyzer is included for generating ultra-short time characteristics by linear prediction coefficients of an input speech signal.
3. The language processing system of MEMS and the method thereof according to claim 1, wherein: the initial pulse of each of the series of equal amplitude, variable sign, variable spacing bursts is located at the same position.
4. The language processing system of MEMS and the method thereof according to claim 1, wherein: the target vector unit comprises a global decision determiner comprising an perceptual weighting filter for filtering the constant amplitude, variable sign, variable spacing pulse sequences.
5. The language processing system of MEMS and the method thereof according to claim 2, wherein: the ultra-short time analyzer has the functions and characteristics of combining MEMS recognition sensing and conversion.
6. The language processing system of MEMS and method thereof according to claims 1-5, wherein: the MEMS recognition sensing unit comprises an input end, a target vector unit and a sensing output end, wherein the input end of the MEMS recognition sensing unit inputs a voice signal and can also generate long-term characteristics, the long-term characteristics at least comprise a tone value of the input voice signal and short-term characteristics of the MEMS recognition sensing output end, and the target vector unit is used for generating a target vector at least by the input voice signal and optional short-term and long-term characteristics.
7. The language processing system of MEMS and the method thereof according to claim 6, wherein: the initial pulses of the pulse sequence are located at the same sample position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810607268.XA CN110660396A (en) | 2018-06-13 | 2018-06-13 | Language processing system and method based on MEMS |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810607268.XA CN110660396A (en) | 2018-06-13 | 2018-06-13 | Language processing system and method based on MEMS |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110660396A true CN110660396A (en) | 2020-01-07 |
Family
ID=69026169
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810607268.XA Pending CN110660396A (en) | 2018-06-13 | 2018-06-13 | Language processing system and method based on MEMS |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110660396A (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1995030222A1 (en) * | 1994-04-29 | 1995-11-09 | Sherman, Jonathan, Edward | A multi-pulse analysis speech processing system and method |
-
2018
- 2018-06-13 CN CN201810607268.XA patent/CN110660396A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1995030222A1 (en) * | 1994-04-29 | 1995-11-09 | Sherman, Jonathan, Edward | A multi-pulse analysis speech processing system and method |
| CN1153566A (en) * | 1994-04-29 | 1997-07-02 | 乔纳森·爱德华·谢尔曼 | Multi-pulse analysis speech processing system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10930270B2 (en) | Processing audio waveforms | |
| CN110827837B (en) | Whale activity audio classification method based on deep learning | |
| CN106782501B (en) | Speech feature extraction method and device based on artificial intelligence | |
| Gevaert et al. | Neural networks used for speech recognition | |
| US8924209B2 (en) | Identifying spoken commands by templates of ordered voiced and unvoiced sound intervals | |
| CN106653056B (en) | Fundamental frequency extraction model and training method based on LSTM recurrent neural network | |
| US9454976B2 (en) | Efficient discrimination of voiced and unvoiced sounds | |
| CN106297826A (en) | Speech emotional identification system and method | |
| RU2016127609A (en) | METHOD AND DEVICE FOR DETERMINING SIGNALS IN DATA | |
| JPS62231997A (en) | Voice recognition system and method | |
| CN106098079B (en) | Method and device for extracting audio signal | |
| Rudoy et al. | Time-varying autoregressions in speech: Detection theory and applications | |
| Tsenov et al. | Speech recognition using neural networks | |
| CN105336344A (en) | Noise detection method and apparatus thereof | |
| CN104732984B (en) | A kind of method and system of quick detection single-frequency prompt tone | |
| CN110660396A (en) | Language processing system and method based on MEMS | |
| JP2008170789A (en) | Voice section detection apparatus and voice section detection method | |
| Raczynski | Speech processing algorithm for isolated words recognition | |
| Teacher et al. | Experimental, limited vocabulary, speech recognizer | |
| CN107039046B (en) | Voice sound effect mode detection method based on feature fusion | |
| Zhao et al. | Online noise estimation using stochastic-gain HMM for speech enhancement | |
| Ho et al. | Naaloss: Rethinking the objective of speech enhancement | |
| US20150364146A1 (en) | Method for Providing Visual Feedback for Vowel Quality | |
| Thakur et al. | Speech enhancement using Open-Unmix music source separation architecture | |
| KR101647059B1 (en) | Independent vector analysis followed by HMM-based feature enhancement for robust speech recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200107 |
|
| RJ01 | Rejection of invention patent application after publication |