Unlock instant, AI-driven research and patent intelligence for your innovation.

Voice recognition server, telephone equipment, voice recognition system, and voice recognition method

A technology of voice recognition and server, which is applied in voice recognition, telephone communication, detailed information of telephone users, etc., can solve the problems of voice recognition performance degradation and model precision degradation, and achieve the effects of reducing capacity, improving recognition accuracy, and improving performance

Inactive Publication Date: 2010-08-04
NTT DOCOMO INC
View PDF1 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

That is, although the utterances from the same terminal are usually the voice of the same user, voice recognition is performed according to different models for each different number, and each different model is updated individually, which may reduce the accuracy of the model, Reduced performance of voice recognition

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Voice recognition server, telephone equipment, voice recognition system, and voice recognition method
  • Voice recognition server, telephone equipment, voice recognition system, and voice recognition method
  • Voice recognition server, telephone equipment, voice recognition system, and voice recognition method

Examples

Experimental program
Comparison scheme
Effect test

no. 1 Embodiment approach

[0055] (Overall structure of voice recognition system 1)

[0056] First, refer to figure 1 as well as figure 2 The overall configuration of the voice recognition system 1 according to the first embodiment of the present invention will be described. figure 1 as well as figure 2 It is a schematic configuration diagram of the voice recognition system 1 . like figure 1 As shown, the voice recognition system 1 is composed of a telephone 100 and a voice recognition server 200 , and the telephone 100 and the voice recognition server 200 are connected to each other through a communication network 300 . The telephone 100 is a mobile telephone capable of using a plurality of telephone numbers and mail addresses (called "two-in-one service" in Japan) in one terminal. Voice recognition server 200 is a server device that converts voice from telephone 100 into characters and transmits the result to telephone 100 . Additionally, if figure 2 As shown, the voice recognition system 1 ...

no. 2 Embodiment approach

[0092] Next, a second embodiment of the present invention will be described. In addition, the description of the part which overlaps with the said 1st Embodiment is abbreviate|omitted, and it demonstrates centering on the difference from 1st Embodiment.

[0093] Figure 9 It is a schematic configuration diagram of the voice recognition server 250 of the second embodiment. Compared with the voice recognition server 200 in the first embodiment, the voice recognition server 250 further includes: a number conversion data storage unit 214 (corresponding to the “data storage unit” in the claims), a number conversion unit 216 (corresponding to the “data storage unit” in the claims) The "model selection unit" in the claims) and the number control unit 218 (equivalent to the "correspondence control unit" in the claims).

[0094] The number conversion data storage unit 214 stores a plurality of telephone numbers available for one telephone in association with user identification infor...

no. 3 Embodiment approach

[0107] Next, a third embodiment of the present invention will be described. In addition, the description of the parts overlapping with the first embodiment described above will be omitted, and the differences from the first embodiment will be mainly described.

[0108] Figure 15 It is a schematic configuration diagram of the voice recognition server 260 of the third embodiment. Compared with the voice recognition server 200 in the first embodiment, the voice recognition server 260 further includes a pattern recognition information receiving unit 220 (corresponding to "voice receiving means" in the claims). The pattern recognition information receiving unit 220 receives pattern recognition information. The pattern recognition information is information that the model selection unit 206 refers to in order to select an acoustic model and a language model. The pattern indicated by the pattern recognition information may specify, for example, a telephone number that can be used...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

A voice recognition server 200 has a voice reception unit 202 which receives a voice from a telephone equipment 100, a model storage unit 208 which stores at least one acoustic model and at least one language model used for converting the voice received by the voice reception unit 202, to character data, a number decision unit 204 which decides a current calling number and a second number of the telephone equipment 100, a model selection unit 206 which selects an acoustic model stored in the model storage unit 208, based on the current calling number and the second number, and which selects a language model stored in the model storage unit 208, based on the current calling number, and a voice recognition unit 210 which converts the voice received by the voice reception unit 202, to character data, based on the acoustic model and the language model selected by the model selection unit 206.

Description

technical field [0001] The invention relates to a voice recognition server, a telephone set, a voice recognition system and a voice recognition method. Background technique [0002] Conventionally, for example, as disclosed in Patent Document 1, there is known a technique in which a dictionary for voice recognition is switched according to a telephone number when voice recognition is performed on a voice uttered by a user. Also, as disclosed in Non-Patent Document 1, for example, a service (so-called 2-in-1 (2in1) service) that allows multiple telephone numbers and mail addresses to be used on one terminal has been developed. [0003] [Patent Document 1] Japanese Patent Laid-Open No. 2000-10590 [0004] [Non-Patent Document 1] Development of 2in1 Service System (System Development of 2-in-1 Service), NTT DoCoMo Technical·Jiyananal, vol.15No.3, P11-19 [0005] In the service using the above-mentioned prior art, when a plurality of numbers are used on the same terminal and d...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G10L15/28G10L15/06G10L15/00G10L15/30
CPCH04M2201/405H04M2250/66H04M3/42153G10L2015/228H04M2201/40G10L15/183G10L15/30
Inventor 张志鹏古川博崇
Owner NTT DOCOMO INC