EP0282272A1
Expired / Lapsed
EP
⏱️ 0.0 years remaining
Voice recognition system
Assignee: Fujitsu Ltd
•
Filing: 1988-03-09
•
Granted: —
•
Expires: 2008-03-09
Abstract
Abstract A voice recognition system selecting word templates necessary for voice recognition from among a plurality of word templates, including: an input phoneme extraction unit (31) for extracting distinctive phonemes which can be clearly recognized from an input pattern of the voice, and all phonemes possibly existing in the input pattern; a dictionary phoneme extraction unit (32) for extracting the distinctive phonemes which can be clearly recognized and can be previously determined at each word template, and all phonemes possibly existing at each word template; and a pre-selector (33) for selecting valid word templates for performing voice recognition based on phoneme information extracted by the input phoneme extraction unit (31) and the dictionary phoneme extraction unit (32). Such a system can enable words to be recognised at high speed from among a large vocabulary.
Inventors
- 👤 Shinta Kimura
- 👤 Toru Sanada
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-11-07 21:08
CPC Classifications
G; G10; G10L; G10L15/00; G10L15/08; G10L15/10; G10L15/28; G10L15/02; G10L2015/025; G10L15/22; G10L2015/226; G10L2015/228
Claims Analysis
Total Claims: 7
Claim 1
Independent Claim
110 words
A voice recognition system selecting word templates necessary for voice recognition from among a plurality of word templates, comprising: an input phoneme extraction unit for extracting distinctive phonemes which can be clearly recognized from an input pattern of the voice, and all phonemes possibly existing in the input pattern; a dictionary phoneme extraction unit for extracting the distinctive phonemes which are clearly recognized and can be previously determined at each word template, and all phonemes possibly existing at each word template; and a pre-selector for selecting valid word templates for performing voice recognition based on phoneme information extracted by said input phoneme extraction unit and said dictionary phoneme extraction unit.
Claim 2
Dependent Claim
110 words
A voice recognition system selecting word templates necessary for voice recognition from among a plurality of word templates, comprising: an input phoneme extraction unit for extracting distinctive phonemes which can be clearly recognized from an input pattern of the voice, and all phonemes possibly existing in the input pattern; a dictionary phoneme extraction unit for extracting the distinctive phonemes which are clearly recognized and can be previously determined at each word template, and all phonemes possibly existing at each word template; and a pre-selector for selecting valid word templates for performing voice recognition based on phoneme information extracted by said input phoneme extraction unit and said dictionary phoneme extraction unit.
Claim 3
Dependent Claim
50 words
A voice recognition system as claimed in claim 1, wherein said input phoneme extraction unit generates bit expression values (I') showing the distinctive phonemes which can be clearly recognized from the input pattern, and generates the other bit expression values (I) showing all phonemes possibly existing in the input pattern.
Claim 4
Dependent Claim
65 words
A voice recognition system as claimed in claim 1, wherein said dictionary phoneme extraction unit generates bit expression values (D i ') showing the distinctive phonemes which can be clearly recognized and can be previously determined at each word template, and generates other bit expression values (D;) showing all phonemes possibly existing in each word template, where, i shows a number of the word template.
Claim 5
Dependent Claim
43 words
A voice recognition system as claimed in any of claims 1 to 3, wherein said pre-selector selects valid word templates by calculating the following formula, that is, and by determining whether or not all bits of the above formula indicate the value "1".
Claim 6
Dependent Claim
107 words
A voice recognition system as claimed in claim 1, wherein said input phoneme extraction unit generates bit expression values (I') showing the distinctive phonemes which can be clearly recognized from the input pattern, said dictionary phoneme extraction unit generates bit expression value (D;') showing the distinctive phonemes which can be clearly recognized and can be previously determined at each word template and generates other bit expression values (D;) showing all phonemes possibly existing in each word template, and said pre-selector selects valid word templates by calculating the following formula, that is, and by judging whether or not all bits of the above formula indicate the value "1".
Claim 7
Dependent Claim
50 words
A voice recognition system as claimed in any of claims 1 to 5, further comprising a bit expression memory between said dictionary phoneme extraction unit and said pre-selector, said bit expression memory storing the bit expression values (D; and D i ') previously calculated by said dictionary phoneme extraction unit.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.