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

A speech recognition effect automatic test system and test method

A technology for automated testing and speech recognition, applied in speech recognition, speech analysis, instruments, etc., can solve problems such as inability to determine the achievement of system effects, lack of recognition effect evaluation methods, and lack of verification of results, so as to solve the problem that test results cannot be repeated. The effect of existing problems, improving test efficiency and accuracy, and reducing human participation

Active Publication Date: 2016-10-19
讯飞医疗科技股份有限公司
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The patent lacks an evaluation method for the recognition effect, only describes the technical implementation, lacks the verification of the results, and cannot determine the achievement of the system effect achieved in this way

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
  • A speech recognition effect automatic test system and test method
  • A speech recognition effect automatic test system and test method
  • A speech recognition effect automatic test system and test method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] In order to better understand the present invention, some basic concepts are firstly explained.

[0044] Open dictation: also known as continuous speech recognition, it is a recognition technology mainly used to recognize large segments of audio content, and can be widely used in text message dictation, voice chat, Weibo, etc.

[0045] Command word recognition: specifically refers to the speech recognition of some commands, such as performing operations on the computer, such as turning on, shutting down, opening Notepad, opening a browser, etc.

[0046] WER: word error rate, mainly for open dictation, because it is basically unfeasible to require sentence accuracy, so this indicator is used to measure it. The calculation formula is: 1-word accuracy. The formula for word accuracy rate is: (number of correct words recognized - number of wrong words inserted) / total number of recognized speech words*100%. The so-called inserting the wrong number of words means adding a re...

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 automatic voice recognition effect testing system and an automatic voice recognition effect testing method. The automatic voice recognition effect testing system comprises an automatic configuration module, an automatic voice playing and recording module and a voice recognition and result statistic module. An operator only needs to prepare testing materials in advance, and the whole execution process is automatic, so that participation of persons is greatly reduced, the labor cost is reduced, repeated test and verification can be performed, large-scale test of few days and nights can be met, automatic statistic is performed on the testing result, manual error is avoided, and the testing efficiency and accuracy can be greatly improved. The system can be applied to equipment such as any desktop computer or notebook computer, and can be used for rapidly obtaining result data in terms of recognition effect evaluation, undertaking tasks which can not be finished by manpower and providing great convenience to each manufacturer.

Description

technical field [0001] The invention relates to a voice recognition effect automatic test system and a test method, belonging to the voice recognition technology. Background technique [0002] Speech recognition research began in the 1950s, initially to convert natural human speech into computer-readable input. [0003] The speech recognition system basically includes speech signal preprocessing and feature extraction, acoustic comparison, and language decoding. Feature extraction is to analyze and process the speech signal, properly remove redundant information that has nothing to do with speech recognition, and obtain important information that affects speech recognition. Acoustic contrast is the pattern matching of speech features to an acoustic model produced using a training algorithm. Language decoding is to use the grammatical network composed of recognized speech commands or the language model composed of statistical methods to perform grammatical and semantic anal...

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): G10L25/60G10L25/69G10L15/18G10L15/28
Inventor 程美鹿晓亮吴如伟朱明凤吴晓杰
Owner 讯飞医疗科技股份有限公司
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com