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

High-performance speech recognition co-processor and method thereof for realizing co-processing

A speech recognition and coprocessor technology, applied in speech recognition, speech analysis, instruments, etc., to improve performance, reduce power consumption, and reduce costs

Inactive Publication Date: 2009-12-02
北京凌声芯语音科技有限公司
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] In order to achieve real-time processing of the speech recognition system, the existing technology can only use DSP to calculate the output probability matrix

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
  • High-performance speech recognition co-processor and method thereof for realizing co-processing
  • High-performance speech recognition co-processor and method thereof for realizing co-processing
  • High-performance speech recognition co-processor and method thereof for realizing co-processing

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] The specific embodiment of the present invention will be further described in detail below in conjunction with the accompanying drawings. The above and other objects, features and advantages of the present invention will be apparent to those skilled in the art from the detailed description of the present invention.

[0031] This embodiment provides a high-performance speech recognition coprocessor, which can be used in cooperation with various microcontrollers MCU to complete the operation of the output probability matrix based on the HMM speech recognition algorithm.

[0032] see image 3 , the coprocessor includes a storage module and an output probability calculation module. The storage module is composed of the following parts: a first storage module 31 , a second storage module 32 , a third storage module 33 and a fourth storage module 34 . The output probability calculation module is composed of the following parts: a Mahalanobis distance calculation module 40 a...

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 relates to a high-performance speech recognition co-processor and a method thereof for realizing co-processing. The high-performance speech recognition co-processor comprises a storage module and an output probability calculation module; wherein the storage module is used for storing a characteristic vector, a model state mean value vector, a model state covariance vector and a probability matrix obtained by calculation in a calculation process. The output probability calculation module comprises a mahalanobis distance calculation module and a logarithm domain addition calculation module, wherein the mahalanobis distance calculation module calculates a mahalanobis distance according to the data of the characteristic vector, the model state mean value vector and the model state covariance vector. The logarithm domain addition calculation module carries out logarithm domain addition calculation according to a calculation result of the mahalanobis distance calculation module, writes the probability matrix obtained by calculation into the storage modules for storage and outputs the probability matrix to an external processor. The co-processor and the method thereof for realizing co-processing, which are provided by the invention, can carry out output probability matrix calculation based on an HMM speech recognition algorithm, aim at the application of an embedded type speech recognition system and have the advantages of performance increase, cost reduction and power consumption reduction.

Description

technical field [0001] The present invention relates to the field of VLSI SOC (System On Chip, System on Chip) design in the field of information technology, and in particular to the application field of a new generation of embedded speech recognition, specifically a HMM (Hidden Markov Models, Hidden Markov Models) based A high-performance speech recognition coprocessor for the output probability matrix operation of the husband model) and an implementation method of the coprocessor. Background technique [0002] The research of embedded speech recognition system has always been an important development direction of the application of speech recognition technology. It provides good human-computer voice interaction for mobile and miniaturized products, such as the third generation of intelligent language learning machines, voice dialing of telephones, voice retrieval of entertainment products such as mp3, vehicle voice control systems, smart toys and home appliances Voice rem...

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): G10L15/14G10L15/28
Inventor 梁维谦董明李鹏刘志杨华中
Owner 北京凌声芯语音科技有限公司
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