Chinese speech recognizing system and method thereof

A technology of speech recognition and Chinese, applied in speech recognition, speech analysis, instruments, etc., to achieve the effect of simplifying the business control process, reducing the voice interaction process, and saving system maintenance costs

Inactive Publication Date: 2008-12-31
上海言海网络信息技术有限公司
View PDF0 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although foreign speech recognition technology is very mature, there are still many deficiencies, such as Chinese speech recognition, the processing of Chinese language characteristics, such as natural language, keywords, polyphonic characters, etc., and special interactive voice requirements, etc. There are still many deficiencies and deficiencies

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
  • Chinese speech recognizing system and method thereof
  • Chinese speech recognizing system and method thereof
  • Chinese speech recognizing system and method thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0040] see figure 1 , The invention discloses a Chinese speech recognition system, comprising a speech data receiving unit, a digital-to-analog conversion unit, a speech data recognition unit, and a business process control unit connected in sequence.

[0041] The voice data recognition unit includes a voice database, a similar sound recognition module, a fuzzy recognition module, a keyword recognition module, a natural speech recognition module, and a synonym recognition module. The voice database is used to preserve the data information of the sampled voice; the similar sound recognition module compares the data information of the received voice with the data information in the database, returns the credibility of the result, if the credibility exceeds the set value, then This recognition is successful; when the recognition is not successful, select some voices with high recognition rate for the user to choose, or prompt the user to re-pronounce; the keyword recognition modu...

Embodiment 2

[0055] Example 2 - Natural Language Understanding

[0056] General speech recognition applications are based on dialogue, for example, flight inquiries. The user calls to inquire about flight information. In general, the system will ask the day of the week, the flight in the morning, afternoon or evening, the departure city, and the big city. The user finally got the query result, but this process is too cumbersome and time-consuming. If the user can say a sentence, such as "query the flight from Beijing to Shanghai on Friday morning", the system can immediately broadcast the query result to the user. The realization of this convenient and simple process will undoubtedly greatly improve the user's efficiency. This invention realizes natural language understanding through technologies such as keyword retrieval and fuzzy matching. The process and results are as follows: image 3 shown.

Embodiment 3

[0057] Embodiment 3 - Similar Sound Recognition

[0058] There are many similar sounds in Chinese. Take an example of an automatic voice switchboard. A company has multiple employees whose names are similar or close in pronunciation. For example, there is a male employee named "Li Xiang", a female employee named "Li Xiang", and others such as Li Qiang , Li Xiang and so on. If the user looks for Li Xiang, the system finds that the recognition results of Li Xiang and Li Xiang are very close, both exceeding the empirical value (such as 85). Based on this situation, the application process cannot determine the user's choice after receiving the result, but will further prompt the user. Whether Li Xiang is male or Li Xiang is female, if the user says Li Xiang is male, the system will easily determine the recognition result and complete the user's operation, such as Figure 4 shown. Or ask the user whether it is "Li Xiang" or "Li Xiang", for the user to choose.

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 Chinese speech recognition system which comprises a speech data reception unit, a digital to analog conversion unit, a speech data recognition unit and a business flow control unit in sequence, wherein, the speech data recognition unit comprises a speech database, a similar sound recognition module, and a fuzzy recognition module; the business flow control unit is used for controlling business flow according to the received speech information from the speech data recognition unit. The system effectively improves Chinese speech recognition rate by the development of speech recognition technology, such as 'key word recognition system', 'nickname, synonym and parasynonyms recognition', and 'fuzzy recognition', etc.

Description

technical field [0001] The invention belongs to the technical field of computer intelligence applications, and relates to a speech recognition system, in particular to a Chinese speech recognition system; meanwhile, it also relates to a speech recognition method of the above-mentioned Chinese speech recognition system. Background technique [0002] Speech recognition technology was researched earlier in foreign countries, and the technology is relatively mature. It has been applied in telecommunications, finance and other industries since the 1990s. Well-known speech recognition products, such as the Nuance speech recognition system, use a client / server structure and can run on Windows and many UNIX systems. It can be flexibly expanded and supports small to large applications. The Nuance system can support large-capacity speech retrieval, natural language recognition, etc., and has supported speech recognition in more than 20 languages ​​including Chinese. Although foreign ...

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): G10L15/28
Inventor 陈洪林
Owner 上海言海网络信息技术有限公司
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