FIELD: electronics.
SUBSTANCE: invention relates to speech recognition. Speech recognition method comprises steps of: receiving a speech signal, processing speech signal, selecting, in normalised spectrum, pauses, noise and sound signals, identifying and converting speech signal, determining presence/absence therein of acoustic features of speech signal, determining probability of all phoneme conditions, processing recognition hypothesis, comparing parameters of recognition hypothesis, syntactical matching of hypothesis, generating recognition result, converting recognition results of all segments of speech signal, outputting speech signal in form of connected text.
EFFECT: technical result is reducing amount of memory elements, needed to store a preset dictionary, and reduced complexity of computer recognition process.
1 cl, 5 dwg, 2 tbl
| Title | Year | Author | Number |
|---|---|---|---|
| SYSTEM AND METHOD OF SPEECH RECOGNITION | 2011 |
|
RU2466468C1 |
| METHOD AND DEVICE FOR CLASSIFYING NOISY VOICE SEGMENTS USING MULTISPECTRAL ANALYSIS | 2014 |
|
RU2606566C2 |
| METHOD AND SYSTEM FOR LEXICAL INTERPRETATION OF FUSED SPEECH | 1997 |
|
RU2119196C1 |
| COMPUTER EQUIPMENT FOR READING OF PRINTED TEXT | 1996 |
|
RU2113726C1 |
| METHOD FOR PRELIMINARY PROCESSING OF TEXT | 2007 |
|
RU2386178C2 |
| METHOD FOR RECOGNIZING WORDS IN CONTINUOUS SPEECH | 2005 |
|
RU2297676C2 |
| METHOD OF RECOGNISING KEY WORDS IN CONTINUOUS SPEECH | 2008 |
|
RU2403628C2 |
| METHOD FOR RECOGNITION OF WORDS IN CONTINUOUS SPEECH AND DEVICE WHICH IMPLEMENTS SAID METHOD | 1996 |
|
RU2101782C1 |
| SYSTEM AND METHOD OF CONVERTING VOICE SIGNAL INTO TRANSCRIPT PRESENTATION WITH METADATA | 2014 |
|
RU2589851C2 |
| METHOD OF RECOGNITION OF SEPARATE WORDS OF SPEECH WITH ADAPTATION TO ANNOUNCER | 1994 |
|
RU2047912C1 |
