Speech recognition system
First Claim
Patent Images
1. In a speech recognition system having input means for inputting a speaker'"'"'s words;
- analyzing means for analyzing and converting the distinctive features of that speaker'"'"'s method of speech into a distinctive features pattern;
memory means for storing the distinctive features; and
comparison means for comparing voice patterns to be recognized with previously stored reference patterns,an improvement comprising;
memory means for temporarily storing the input voice data and its time interval and;
reproduction means controlled by the time interval for audibly outputting the input voice data so that the speaker can determine whether his intended speech was correctly patterned and stored timewise by the recognition system.
1 Assignment
0 Petitions
Accused Products
Abstract
In a speech analysis/recognition system, the input utterance is also directly registered (stored) for later readout verification of correct start and end control signals.
-
Citations
3 Claims
-
1. In a speech recognition system having input means for inputting a speaker'"'"'s words;
- analyzing means for analyzing and converting the distinctive features of that speaker'"'"'s method of speech into a distinctive features pattern;
memory means for storing the distinctive features; and
comparison means for comparing voice patterns to be recognized with previously stored reference patterns,an improvement comprising;
memory means for temporarily storing the input voice data and its time interval and;
reproduction means controlled by the time interval for audibly outputting the input voice data so that the speaker can determine whether his intended speech was correctly patterned and stored timewise by the recognition system. - View Dependent Claims (2, 3)
- analyzing means for analyzing and converting the distinctive features of that speaker'"'"'s method of speech into a distinctive features pattern;
Specification