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

Distributed speech recognition and retrieval method

A speech recognition and distributed technology, applied in speech recognition, speech analysis, instruments, etc., can solve the problems of lower recognition rate, speech distortion, large storage space, etc., to reduce the space occupied, improve the accuracy, and facilitate the use Effect

Inactive Publication Date: 2017-12-29
北京新岸线网络技术有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The disadvantage is that the calculation area is large, a large storage space is required, the transmission occupies a large channel bandwidth, and the voice is disturbed by noise during transmission, which will cause distortion and reduce the recognition rate.

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
  • Distributed speech recognition and retrieval method
  • Distributed speech recognition and retrieval method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0052] A kind of distributed speech recognition and retrieval method provided by the present invention, refer to figure 1 , the method includes:

[0053] Step S101, a voice recognition unit is set on the server side, and the voice recognition unit creates a voice configuration file for each registered user;

[0054] Step S102, the user logs in the browser of the client to retrieve the voice, and the voice information is transmitted to the server through the client;

[0055] Step S103, the server recognizes the voice according to the voice recognition engine and the user's configuration file, and returns the retrieval result to the client.

Embodiment 2

[0057] Step S101, a voice recognition unit is set on the server side, and the voice recognition unit creates a voice configuration file for each registered user;

[0058] In some optional embodiments, for non-registered users, the server automatically invokes a region-specific voice recognition configuration file for recognition according to the user's IP.

[0059] Step S102, the user logs in the browser of the client to retrieve the voice, and the voice information is transmitted to the server through the client;

[0060] Step S103, the server recognizes the voice according to the voice recognition engine and the user's configuration file, and returns the retrieval result to the client.

[0061] Preferably, the server includes one or more of the following retrieval units:

[0062]The voice pre-storage retrieval unit is used to analyze, save, retrieve and maintain commonly used hot voice information in advance;

[0063] A voice cache retrieval unit, configured to retrieve th...

Embodiment 3

[0073] Such as figure 2 Shown, a kind of speech retrieval method that the present invention also provides, comprises:

[0074] Step S201, the client receives the original audio;

[0075] Step S202, preprocessing the original audio to obtain audio metadata;

[0076] Step S203, according to different query conditions input by the user, use relevant metadata in the audio metadata to retrieve the desired content;

[0077] Step S204, saving frequently used hotspot voice information and hotspot voice information with the most searches in the database.

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 method for distributed speech recognition and retrieval, which includes: setting a speech recognition unit at the server end, and the speech recognition unit establishes a speech configuration file for each registered user; The voice information is sent to the server through the client; the server recognizes the voice according to the voice recognition engine and the user's configuration file, and returns the retrieval result to the client. By adopting the technical solution provided by the invention, the space occupied by the voice recognition client is reduced, the user is more convenient to use, and the correct rate of voice recognition is improved.

Description

technical field [0001] The invention relates to the field of voice retrieval based on content, in particular to a distributed voice recognition and retrieval method. Background technique [0002] Traditional speech recognition devices mostly use centralized speech recognition technology, and tasks are mainly completed on a single device. This method faces the complex network environment, and the precise retrieval of resources has become the general trend. The continuous development of mobile networks makes handheld retrieval possible. The disadvantage is that the calculation area is large, a large storage space is required, the transmission occupies a large channel bandwidth, and the voice is disturbed by noise during transmission, which will cause distortion and reduce the recognition rate. In the current Internet era, it is unrealistic to use centralized speech recognition on various mobile devices. Therefore, distributed speech recognition technology is required. This te...

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): G10L15/08G10L15/30
CPCG10L15/08G10L15/30
Inventor 罗旻白永强鲍东山
Owner 北京新岸线网络技术有限公司