Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Entity error correction method and system for voice transcription text

A technology of speech transcription and error correction method, which is applied in speech analysis, speech recognition, natural language data processing, etc., and can solve the problem of low accuracy

Active Publication Date: 2020-09-01
GLOBAL ENERGY INTERCONNECTION RES INST CO LTD +3
View PDF10 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of this, the embodiment of the present invention provides a method and system for entity error correction of voice transcription text to overcome the problem of low accuracy of the entity error correction method for voice transcription text in the prior 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
  • Entity error correction method and system for voice transcription text
  • Entity error correction method and system for voice transcription text
  • Entity error correction method and system for voice transcription text

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0029] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present invention.

[0030] The technical features involved in different embodiments of the present invention described below may be combined with each other as long as they do not constitute a conflict with each other.

[0031] When all kinds of voice content stored in the power grid system are automatically transcribed into text, due to the influence of accents, sentence breaks...

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 provides an entity error correction method and system for a voice transcription text. The method comprises the following steps: carrying out pinyin labeling on entity vocabularies extracted from a target voice transcription text; clustering the entity vocabularies by utilizing the marked pinyin and an editing distance based on pinyin similarity to generate a clustering result; and determining the entity vocabulary with the highest occurrence frequency in the same category in the clustering result as a standard entity vocabulary, and replacing other entity vocabularies in the category with the standard entity vocabulary. Entity vocabularies are clustered by utilizing an editing distance based on pinyin similarity; the pinyin similarity is taken as a reference factor and the pinyin similarity is added into an editing distance algorithm; the ability of distinguishing synonyms and phonetic words is enhanced, the clustering result better conforms to the actual situation of thevoice transcription text, other entity vocabularies are replaced with the entity vocabularies with the highest occurrence frequency in the same category according to the clustering result, error correction of the voice transcription text is achieved, and then the accuracy of the final voice transcription text is improved.

Description

technical field [0001] The invention relates to the technical field of speech processing, in particular to a method and system for entity error correction of speech transcription text. Background technique [0002] With the promotion and deepening of artificial intelligence (AI, Artificial Intelligence) technology, a batch of intelligent products represented by live working robots and AI controllers have taken the lead in entering the electric power industry and exerted great effectiveness. Therefore, it is the current development trend to combine artificial intelligence technology with power, energy and other industries to promote the transformation and upgrading of traditional industries. Speech is the most natural and effective way for humans to communicate, making speech recognition technology a popular research direction. At present, a large number of call records are generated in the State Grid customer service center every day. The speech of these calls is automatica...

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): G06F40/232G06F40/295G10L15/26
CPCG06F40/232G06F40/295G10L15/26
Inventor 贾全烨张强宋博川柴博
Owner GLOBAL ENERGY INTERCONNECTION RES INST CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products