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

Voice age recognition method and system thereof

A recognition method and recognition system technology, applied in the field of speech age recognition, can solve the problems of low data labeling efficiency and increased labor costs, and achieve the effects of improving speech recognition rate, reducing labor costs, and improving efficiency

Inactive Publication Date: 2019-03-08
北京哈沃在线科技有限公司
View PDF3 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In order to improve the recognition rate, we can only clean the training data. The current method of cleaning according to business needs is to manually filter adult voices and select children's voices, which leads to low efficiency of data labeling and increases labor costs.

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 age recognition method and system thereof
  • Voice age recognition method and system thereof
  • Voice age recognition method and system thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0022] A voice age recognition method such as figure 1 As shown, it mainly includes the following steps:

[0023] Step S101, collecting speech data to be recognized;

[0024] In this embodiment, in the application of the actual students' class process, as the course progresses, the students answer the corresponding questions according to the teacher's questions. At this time, the audio collection module of the online education client constantly collects the speech data (PCM data) to be recognized, and converts the PCM data Send it to the SAR (Speech Age Recognition) module.

[0025] Step S102, performing preprocessing on the speech data to be recognized to obtain first preprocessing data;

[0026] In step S102, if figure 2 As shown, after the SAR module receives the PCM data, it judges whether the SAR module is enabled for the first time. If so, the SVM model is loaded first, and then the speech data to be recognized is preprocessed; otherwise, the speech data to be recog...

Embodiment 2

[0074] A speech age recognition system such as Figure 7As shown, it mainly includes: an audio collection module 401, which is used to collect speech data to be recognized; a preprocessing module 402, which is used to preprocess the speech data to be recognized to obtain the first preprocessed data; a feature extraction module 403, which is used to process the speech data to be recognized Perform feature extraction on the preprocessed data to obtain the first feature parameter; the dimensionality reduction aggregation module 404 is used to perform dimensionality reduction and aggregation on the first feature parameter to obtain the second feature parameter; the SVM identification module 405 is used to obtain the second feature parameter according to the SVM model and The second feature parameter identifies the age group to which the voice data to be recognized belongs.

[0075] In this embodiment, in the application of the actual students' class process, as the course progress...

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 age recognition method and a system thereof. The method includes the following steps: collecting voice data to be identified; performing pre-processing on the voice data to be identified to obtain a first pre-processed data; performing feature extraction on the first pre-processed data to obtain a first feature parameter; performing dimension reduction and aggregation on the first feature parameter to obtain a second feature parameter; loading a pre-stored SVM model, and identifying an age period of the to-be-identified voice data according to the SVM model andthe second feature parameter. Compared with the prior art, the voice age recognition method and the system replace the manual with a computer, and the non-child voice is well filtered, the speed of voice annotation is accelerated, the efficiency of full manual annotation is improved, the speech recognition rate is increased, and the labor cost is reduced.

Description

technical field [0001] The invention relates to a voice age recognition method and system, belonging to the technical field of speaker recognition. Background technique [0002] With the rise of artificial intelligence research, a large number of human-computer interaction products need to distinguish the speaker itself, especially in the online education industry. Accurate and real-time identification of the person attending the class is helpful for intelligent interaction, tracking learning progress, and improving learning efficiency. great help. [0003] At present, in the vertical field of children's English speech recognition, due to data interference in acoustic model training, and most of the data interference comes from adult spoken language, the recognition rate will obviously drop in the process of children's English recognition. In order to improve the recognition rate, we can only clean the training data. The current method of cleaning according to business need...

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): G10L25/51G10L17/04G10L17/02G10L25/03
CPCG10L17/02G10L17/04G10L25/03G10L25/51
Inventor 马金龙杨欣陈靖
Owner 北京哈沃在线科技有限公司