JPH03287200A - Voice recognition device - Google Patents
Voice recognition deviceInfo
- Publication number
- JPH03287200A JPH03287200A JP2088011A JP8801190A JPH03287200A JP H03287200 A JPH03287200 A JP H03287200A JP 2088011 A JP2088011 A JP 2088011A JP 8801190 A JP8801190 A JP 8801190A JP H03287200 A JPH03287200 A JP H03287200A
- Authority
- JP
- Japan
- Prior art keywords
- hypothesis
- symbol sequence
- voice
- end time
- model
- 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
- 108010076504 Protein Sorting Signals Proteins 0.000 abstract 3
- 230000003247 decreasing effect Effects 0.000 abstract 1
- 230000003287 optical effect Effects 0.000 abstract 1
Abstract
PURPOSE:To recognize a voice corresponding to an optical signal sequence efficiently by providing a candidate selecting means which outputs a signal sequence of candidates from a hypothesis table as a voice recognition result. CONSTITUTION:The hypothesis table 21 contains hypotheses consisting of the symbol sequence of an input voice and the end time of a voice section, a hypothesis selecting means 22 selects the hypothesis of the end time, and a hypothesis generating means 23 couples a specific signal with the signal sequence to generate a symbol sequence of the new hypothesis. A model generating means 3 generates a voice model represented by the symbol sequence depending upon the symbol sequence and the phoneme context obtained from the features of the input voice. A hypothesis registration means 41 compares the voice model with the features of the input voice to detect the end time when the likelihood becomes maximum and generates and registers a new hypothesis in the hypothesis table 21. A candidate registering means 5 selects the symbol sequence of candidates and outputs it as the voice recognition result. Consequently, the number of hypotheses which are generated is decreased and a voice corresponding to the optional symbol sequence can be recognized efficiently.
Description
【発明の詳細な説明】
〔産業上の利用分野〕
この発明は入力音声を認識する音声認識装置に関するも
のである。
〔従来の技術〕DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a speech recognition device that recognizes input speech. [Conventional technology]
Claims (1)
説を保持する仮説テーブルと、この仮説テーブルから終
端時刻の仮説を選択する仮説選択手段と、この仮説選択
手段の選択した仮説の記号系列に所定の記号を結合して
新たな仮説の記号系列を生成する仮説生成手段と、この
仮説生成手段の生成した記号系列の表す音声モデルを該
記号系列並びに入力音声の特徴から得られる音素文脈に
依存して生成するモデル生成手段と、このモデル生成手
段の生成する音声モデルと入力音声の特徴とを比較し尤
度が極大となる音声区間の終端時刻を検出しこの終端時
刻に基づき新しい仮説を生成して上記仮説テーブルに登
録する仮説登録手段と、上記仮説テーブルから候補の記
号系列を選択し音声認識結果として出力する候補選択手
段とを備えたことを特徴とする音声認識装置。A hypothesis table that holds a hypothesis consisting of a symbol sequence of an input voice and an end time of a speech interval; a hypothesis selection means for selecting a hypothesis of an end time from this hypothesis table; A hypothesis generating means that combines predetermined symbols to generate a new hypothetical symbol sequence, and a speech model represented by the symbol sequence generated by this hypothesis generating means depends on the phoneme context obtained from the symbol sequence and the characteristics of the input speech. A model generation means that generates a model by comparing the voice model generated by this model generation means with the characteristics of the input voice, detects the end time of the voice section where the likelihood is maximum, and generates a new hypothesis based on this end time. A speech recognition device comprising: hypothesis registration means for registering the selected symbol sequence in the hypothesis table; and candidate selection means for selecting a candidate symbol sequence from the hypothesis table and outputting it as a speech recognition result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2088011A JPH087594B2 (en) | 1990-04-02 | 1990-04-02 | Voice recognizer |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2088011A JPH087594B2 (en) | 1990-04-02 | 1990-04-02 | Voice recognizer |
Publications (2)
Publication Number | Publication Date |
---|---|
JPH03287200A true JPH03287200A (en) | 1991-12-17 |
JPH087594B2 JPH087594B2 (en) | 1996-01-29 |
Family
ID=13930902
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP2088011A Expired - Fee Related JPH087594B2 (en) | 1990-04-02 | 1990-04-02 | Voice recognizer |
Country Status (1)
Country | Link |
---|---|
JP (1) | JPH087594B2 (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5185016B2 (en) * | 2008-08-19 | 2013-04-17 | キヤノン株式会社 | Speech recognition apparatus, control method therefor, and program |
-
1990
- 1990-04-02 JP JP2088011A patent/JPH087594B2/en not_active Expired - Fee Related
Also Published As
Publication number | Publication date |
---|---|
JPH087594B2 (en) | 1996-01-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20220165268A1 (en) | Indicator for voice-based communications | |
US10074369B2 (en) | Voice-based communications | |
US10453449B2 (en) | Indicator for voice-based communications | |
US6067520A (en) | System and method of recognizing continuous mandarin speech utilizing chinese hidden markou models | |
CN109155132A (en) | Speaker verification method and system | |
US10148912B1 (en) | User interface for communications systems | |
JPS6466698A (en) | Voice recognition equipment | |
JP2004518155A (en) | System and method for automatic speech recognition using mapping | |
KR19980070329A (en) | Method and system for speaker independent recognition of user defined phrases | |
JP2019211515A (en) | Voice dialogue system, processing method of the same and program thereof | |
KR20190032557A (en) | Voice-based communication | |
JP2004053742A (en) | Speech recognition device | |
Schuller et al. | Comparing one and two-stage acoustic modeling in the recognition of emotion in speech | |
JP2003163951A (en) | Sound signal recognition system, conversation control system using the sound signal recognition method, and conversation control method | |
JPH07219579A (en) | Speech recognition device | |
JPH03287200A (en) | Voice recognition device | |
Seide et al. | Towards an automated directory information system. | |
JP2820093B2 (en) | Monosyllable recognition device | |
JP2000020089A (en) | Speed recognition method and apparatus therefor as well as voice control system | |
KR100259777B1 (en) | Optimal synthesis unit selection method in text-to-speech system | |
JPH0384600A (en) | Voice recognition device | |
JPH119847A (en) | Talkative parrot sounding device | |
JP2007535692A (en) | System and method for computer recognition and interpretation of arbitrarily spoken characters | |
KR100281582B1 (en) | Speech Recognition Method Using the Recognizer Resource Efficiently | |
JP3881155B2 (en) | Speech recognition method and apparatus |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
LAPS | Cancellation because of no payment of annual fees |