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

Voice data recognition method and device for distinguishing regional accents

一种语音数据、语音识别的技术，应用在语音识别、语音分析、仪器等方向，能够解决耗费大人力物力等问题，达到提高效率、减少人工标注的工作量的效果

Inactive Publication Date: 2014-03-26
BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
View PDF5 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in the prior art, there is no automatic method for distinguishing which are Mandarin speech data and which are regional speech data from these speech data, so that before using the original speech data to train an acoustic model for local dialects, it is first necessary to The original voice data is manually marked with geographical tags, which will consume a lot of manpower and material resources

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 data recognition method and device for distinguishing regional accents
  • Voice data recognition method and device for distinguishing regional accents
  • Voice data recognition method and device for distinguishing regional accents

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0019] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, but not to limit the present invention. In addition, it should be noted that, for the convenience of description, only parts related to the present invention are shown in the drawings but not all content.

[0020] figure 1 and figure 2 A first embodiment of the invention is shown.

[0021] figure 1 It is a flow chart of the speech data recognition method for distinguishing regional accents provided by the first embodiment of the present invention. see figure 1 , the speech data recognition method for distinguishing regional accents comprises:

[0022] S110. Calculate speech recognition confidence and / or signal-to-noise ratio of the speech data, and filter regional speech data from the speech data according to the speech rec...

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 invention discloses a voice data recognition method and device for distinguishing regional accents. The voice data recognition method for distinguishing the regional accents includes the steps that the voice recognition confidence coefficient and / or the signal to noise ratio of voice data is calculated, and regional voice data are screened out of the voice data according to the voice recognition confidence coefficient and / or the signal to noise ratio; the region to which the regional voice data belong is determined according to the regional attribute of the regional voice data. According to the voice data recognition method and device for distinguishing the regional accents, the regional voice data are automatically recognized from a mass of voice data through the confidence coefficient and / or the signal to noise ratio of the voice data, so that manual marking performed on the voice data is avoided, and the efficiency of processing the voice data is improved.

Description

technical field [0001] The invention relates to the technical field of voice data processing, in particular to a voice data recognition method and device for distinguishing regional accents. Background technique [0002] my country has a vast territory, which makes there are many types of dialects in Chinese that contain regional accents. Dialects from different places have different pronunciation characteristics and speech speed characteristics. Therefore, when the same acoustic model is used for speech recognition of dialects in different places, the recognition accuracy may not be high. In order to solve the problem of low recognition accuracy when using a unified acoustic model suitable for Mandarin Chinese for speech recognition in different local dialects, it is a good way to train acoustic models customized for each dialect for different local dialects . [0003] Training an acoustic model requires massive amounts of training data. At present, with the popularity ...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/00G10L15/06
CPCG10L15/06G10L15/005G10L15/01G10L25/90G10L25/84G10L15/02G10L15/07
Inventor 苏丹尹钊
Owner BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD