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

Voice decoding method and device

A speech decoding and speech technology, applied in the field of speech decoding, can solve the problems of slow speed and low precision, and achieve the effect of improving speed and precision and reducing the possibility of wrong clipping.

Active Publication Date: 2012-10-17
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF3 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is to provide a voice decoding method and device to solve the technical defects of slow speed and low precision in voice decoding 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
  • Voice decoding method and device
  • Voice decoding method and device
  • Voice decoding method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0023] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0024] Speech recognition usually includes the following parts: performing front-end processing on the speech, extracting the acoustic features of the speech, and decoding the speech feature stream formed after the feature extraction. The front-end processing of speech includes processing the original speech, partially eliminating the influence of noise and different speakers, so that the processed signal can better reflect the essential characteristics of speech. The most commonly used front-end processing includes endpoint detection and speech enhancement. In the stage of extracting the acoustic features of speech, the more commonly used acoustic features include linear prediction coefficient LPC, cepstral coefficient CEP and so on. There are ma...

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 a voice decoding method and device. The voice decoding method comprises the following steps of: A, obtaining a current voice characteristic frame from a voice characteristic flow to be decoded; B, utilizing the current voice characteristic frame to expand each current decoding path; C, utilizing a voice short-time stable characteristic to select more than one decoding path from each expanded decoding path to be used as the current decoding path; judging whether the voice characteristic flow to be decoded reaches to a final frame or not; if so, determining an optimal decoding path from the current each decoding path to be used as a result for decoding the voice characteristic flow to be decoded; and if not, taking a next frame of the voice characteristic flow to be decoded as the current voice characteristic frame, and returning back to the step B. With the adoption of the manner, the precision of voice decoding is improved.

Description

【Technical field】 [0001] The invention relates to speech recognition technology, in particular to a speech decoding method and device. 【Background technique】 [0002] Using the HMM (Hidden Markov Model, Hidden Markov Model) model for speech recognition has become a mainstream technology in speech recognition. HMM is a statistical model established for the time series structure of the speech signal. It regards the speech signal as a mathematical double stochastic process: one is to use a Markov chain with a finite number of states to simulate the hidden changes in the statistical characteristics of the speech signal. is a stochastic process containing , and the other is a random process of the sequence of observations associated with each state of the Markov chain. When using the HMM model for speech decoding, as the decoding process progresses, the number of decoding paths will increase geometrically. Therefore, in order to reduce the amount of calculation and speed up the ...

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): G10L19/00
Inventor 钱胜
Owner BEIJING BAIDU NETCOM SCI & TECH 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

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