Voice recognition device
First Claim
Patent Images
1. A voice recognition device comprising:
- a sound standard model memory for storing a plurality of sound standard models corresponding respectively to a plurality of languages which are recognition objects;
a voice label memory for storing a plurality of user dictionaries corresponding respectively to a plurality of languages;
a first language switching unit for switching among a plurality of languages for said sound standard model memory to select one of the plurality of languages;
a second language switching unit for switching among the plurality of languages for said voice label memory to select one of the plurality of languages;
a control unit for causing said first language switching unit and said second language switching unit to switch in synchronization with each other;
an acoustic analysis unit for performing an acoustic analysis process on an inputted voice uttered by a user to output a feature vector time series of said voice uttered by the user; and
a voice recognition unit for creating a voice label string for said inputted voice uttered by the user on a basis of both the feature vector time series from said acoustic analysis unit, and data about a sound standard model corresponding to the language for said sound standard model memory which is selected by said first language switching unit while said first language switching unit is caused to switch, and registering said voice label string with the language for said voice label memory which is selected by said second language switching unit in synchronization with said first language switching unit.
1 Assignment
0 Petitions
Accused Products
Abstract
A voice recognition unit is constructed in such a way as to create a voice label string for an inputted voice uttered by a user inputted for each language on the basis of a feature vector time series of the inputted voice uttered by the user and data about a sound standard model, and register the voice label string into a voice label memory 2 while automatically switching among languages for a sound standard model memory 1 used to create the voice label string, and automatically switching among the languages for the voice label memory 2 for holding the created voice label string by using a first language switching unit SW1 and a second language switching unit SW2.
-
Citations
7 Claims
-
1. A voice recognition device comprising:
-
a sound standard model memory for storing a plurality of sound standard models corresponding respectively to a plurality of languages which are recognition objects; a voice label memory for storing a plurality of user dictionaries corresponding respectively to a plurality of languages; a first language switching unit for switching among a plurality of languages for said sound standard model memory to select one of the plurality of languages; a second language switching unit for switching among the plurality of languages for said voice label memory to select one of the plurality of languages; a control unit for causing said first language switching unit and said second language switching unit to switch in synchronization with each other; an acoustic analysis unit for performing an acoustic analysis process on an inputted voice uttered by a user to output a feature vector time series of said voice uttered by the user; and a voice recognition unit for creating a voice label string for said inputted voice uttered by the user on a basis of both the feature vector time series from said acoustic analysis unit, and data about a sound standard model corresponding to the language for said sound standard model memory which is selected by said first language switching unit while said first language switching unit is caused to switch, and registering said voice label string with the language for said voice label memory which is selected by said second language switching unit in synchronization with said first language switching unit. - View Dependent Claims (2, 3)
-
-
4. A voice recognition device comprising:
-
a sound standard model memory for storing a plurality of sound standard models corresponding respectively to a plurality of languages which are recognition objects; a voice label memory for storing a user dictionary; a first language switching unit for switching among a plurality of languages for said sound standard model memory to select one of the plurality of languages; a control unit for switching said first language switching unit to a specified language selection position according to a specified language; a voice data memory for holding a voice uttered by a user; an acoustic analysis unit for performing an acoustic analysis process on an inputted voice uttered by the user or the voice uttered by the user which is read out of said voice data memory to output a feature vector time series of said voice uttered by the user; and a voice recognition unit for creating a voice label for the voice uttered by the user according to both data about a sound standard model corresponding to the language selected by said first language switching unit, and said feature vector time series, and registering said voice label into said voice label memory. - View Dependent Claims (5)
-
-
6. A voice recognition device comprising:
-
a sound standard model memory for storing a plurality of sound standard models corresponding respectively to a plurality of languages which are recognition objects; a voice label memory for storing a user dictionary; a first language switching unit for switching among a plurality of languages for said sound standard model memory to select one of the plurality of languages; a control unit for switching said first language switching unit to a specified language selection position according to a specified language; an acoustic analysis unit for performing an acoustic analysis process on an inputted voice uttered by a user to output a feature vector time series of said voice uttered by the user; a sound data memory for holding the feature vector time series which said acoustic analysis unit acquires by performing the acoustic analysis on the inputted voice uttered by the user; and a voice recognition unit for creating a voice label for the voice uttered by the user according to both data about a sound standard model corresponding to the language selected by said first language switching unit, and the feature vector time series read out of said sound data memory, and registering said voice label into said voice label memory. - View Dependent Claims (7)
-
Specification