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

Speech content recognizing device and speech content recognizing method

A content recognition and voice recognition technology, applied in speech recognition, image data processing, speech analysis, etc., can solve problems such as the decline of voice recognition accuracy and the extraction of characteristic patterns of the speaker.

Inactive Publication Date: 2008-01-23
KYOCERA CORP
View PDF1 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] However, in the above-mentioned conventional voice recognition technology, once the speaker is not near the receiver, the characteristic pattern of the voice of the speaker cannot be extracted well, and there is a problem that the accuracy of voice recognition decreases.

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
  • Speech content recognizing device and speech content recognizing method
  • Speech content recognizing device and speech content recognizing method
  • Speech content recognizing device and speech content recognizing method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] Embodiments of the present invention will be described with reference to the drawings.

[0026] The relevant utterance content recognition device 10 of the present invention is, for example, a computer such as a mobile phone, and includes a CPU 12 , a storage unit 14 , an input unit 15 , and an output unit 19 as shown in FIG. 1 .

[0027] The input unit 15 is a functional unit for inputting external information of the utterance content recognition device 10 to the CPU 12 , and includes a microphone 16 and a camera 18 in the present embodiment. The microphone 16 is, for example, a device capable of collecting sound, such as a directional microphone. The microphone 16 has directivity with respect to the sound collecting direction, and the CPU 12 can control the directivity. Specifically, the CPU 12 can control the directivity of the microphone 16 by controlling its posture. In addition, the microphone 16 converts the collected sound into an electrical signal, and output...

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

The object of the present invention is to suppress low-precision voice recognition due to the fact that the speaker is not in the vicinity of the microphone. The utterance content recognition device (10) is characterized in that it has: a microphone (16) for collecting sound, a camera (18) for capturing an image of the speaker who utters a sound to the microphone (16), and an audio recording based on the above-mentioned collected sound. A voice recognition function unit (132) for voice recognition, and when the captured image does not include an image of the speaker representing at least a part of the speaker, performing voice recognition performed by the voice recognition function unit (132) Restricted recognition / learning judgment unit (142).

Description

technical field [0001] The invention relates to an utterance content recognition device and a utterance content recognition method. Background technique [0002] Voice recognition technology that converts voices into character strings is known. In this voice recognition technology, firstly, a microphone is used to collect the voice of the speaker. Next, the characteristic pattern of the collected sound is extracted. Then, a character string pattern corresponding to the extracted characteristic pattern is output as a recognition result, whereby the voice is replaced with a character string. [0003] In addition, Patent Document 1 describes that such a voice recognition technology is used in combination with a lip recognition technology that performs pattern recognition based on a characteristic pattern of a lip shape of a speaker. [0004] Patent Document 1: Japanese Unexamined Patent Publication No. 6-311220 [0005] However, in the above-mentioned conventional voice rec...

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): G10L15/28G06T7/00G06T7/20G10L11/02G10L15/04G10L15/24G10L21/02G10L25/78
CPCG10L15/24G10L25/78G10L15/28G10L15/20G10L21/02
Inventor 守田空悟
Owner KYOCERA CORP