Speech training data processing method, device and storage medium

A technology of speech training and processing method, which is applied in the field of data processing, can solve the problems of expensive speech materials, achieve extremely low cost and solve expensive effects

Active Publication Date: 2022-04-22
GREE ELECTRIC APPLIANCES INC
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The disclosure provides a processing method, device and storage medium for speech training data to solve the technical problem of expensive speech materials used in training speech recognition models in the related art

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 training data processing method, device and storage medium
  • Speech training data processing method, device and storage medium
  • Speech training data processing method, device and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0045] figure 1 It is a flow chart of a method for processing voice training data shown in an exemplary embodiment of the present disclosure to solve the technical problem in the related art that the voice material used for training a voice recognition model is expensive. Such as figure 1 Shown, the processing method of this speech training data comprises:

[0046] S11, acquiring a video file.

[0047] S12. Extract an audio file synchronized with the video file from the video file.

[0048] S13. Identify the frame images with subtitles in the video file, so as to obtain the time period of the frame images with the same subtitles in the video file.

[0049] S14. Intercept the audio file according to the time period to obtain speech training data, where the speech training data includes the intercepted audio file and corresponding subtitles.

[0050] In step S11, the video file can be obtained from the stored video database, or a preset video download website can be provided...

Embodiment 2

[0061] figure 2 It is a schematic diagram of identifying the same subtitle for a video file according to an exemplary embodiment of the present disclosure. In order to solve the technical problem of expensive speech materials used in training speech recognition models in the related art, the figure 2 As an example, the processing method of the speech training data disclosed in the present disclosure is described:

[0062] First, input a video file; then, strip out the entire audio file corresponding to the video file as a backup, since the technology of extracting synchronous audio from the video file is a known technology, it will not be described in detail in this embodiment; then, Use image recognition technology to identify each frame of the video file, such as figure 2 As shown, the words "Happy Birthday to you" are recognized, and the frequency frame with "Happy Birthday to you" is recognized for the first time at this time as A, and the time point corresponding to ...

Embodiment 3

[0065] image 3 It is a speech training data processing device shown in an exemplary embodiment of the present disclosure to solve the technical problem in the related art that speech materials used for training a speech recognition model are expensive. Such as image 3 As shown, the processing device 300 of the voice training data includes:

[0066] Obtaining module 310, configured to obtain video files;

[0067] An extraction module 320 configured to extract an audio file synchronized with the video file from the video file;

[0068] The identification module 330 is configured to identify the frame images with subtitles in the video file, so as to obtain the time period of the frame images with the same subtitles in the video file;

[0069] The intercepting module 340 is configured to intercept the audio file according to the time period to obtain speech training data, and the speech training data includes the intercepted audio file and corresponding subtitles.

[0070] ...

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 present disclosure relates to the technical field of data processing, in particular to a processing method, device and storage medium for speech training data, which are used to solve the technical problem in the related art that speech materials used for training speech recognition models are expensive. The processing method of the speech training data includes: obtaining a video file; extracting an audio file synchronized with the video file from the video file; identifying frame images with subtitles in the video file to obtain images with the same The time period of the subtitle frame image in the video file; according to the time period, the audio file is intercepted to obtain speech training data, and the speech training data includes the intercepted audio file and corresponding subtitles.

Description

technical field [0001] The present disclosure relates to the technical field of data processing, in particular to a method, device and storage medium for processing voice training data. Background technique [0002] At present, the development of artificial intelligence is in full swing, and the application of speech recognition is also emerging in an endless stream. However, before using this technology for speech recognition, the speech recognition model needs to be trained, and training the speech recognition model requires the use of a large number of speech fragments. Contents of the invention [0003] The present disclosure provides a processing method, device and storage medium for speech training data to solve the technical problem in the related art that speech materials used for training speech recognition models are expensive. [0004] In order to achieve the above purpose, the first aspect of the embodiments of the present disclosure provides a method for proc...

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 Patents(China)
IPC IPC(8): G10L15/06G10L15/04G10L25/57
CPCG10L15/063G10L15/04G10L25/57
Inventor 李泽堃
Owner GREE ELECTRIC APPLIANCES INC
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap