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

Multi-Recognizer Speech Recognition

A speech recognition and speech technology, applied in speech recognition, devices with speech recognition, speech analysis, etc., to achieve the effect of improving confidentiality

Active Publication Date: 2019-10-29
GOOGLE LLC
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Tuning each recognizer to a different dictionary

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
  • Multi-Recognizer Speech Recognition
  • Multi-Recognizer Speech Recognition
  • Multi-Recognizer Speech Recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0013] figure 1 A schematic diagram of an example of a system 100 for performing speech recognition using multiple speech recognizers is shown. exist figure 1 In the example of , user 102 issues a command to mobile device 104 . In this example, mobile device 104 is a cellular telephone (also known as a smartphone) with advanced computing capabilities.

[0014] Mobile device 104 receives input audio, such as speech, provided by user 102 and provides the audio to limited speech recognizer 110 and extended speech recognizer 120 over network 106 , such as the Internet or a cellular data network. Limited speech recognizer 110 and extended speech recognizer 120 perform speech-to-text transcription of user 102 utterances. In this example, mobile device 104 may include an application ("APP") that receives incoming audio. The APP may have any suitable functionality, for example it may be a search APP, a messaging APP, an email APP, or the like. In this regard, APP is used as an ex...

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 subject matter of the present specification can be embodied as a method or the like comprising receiving audio data corresponding to an utterance, obtaining a first transcription of the utterance produced using a limited speech recognizer. The limited speech recognizer includes a speech recognizer that includes a language model trained with a limited speech recognition vocabulary that includes one or more terms from the speech command grammar but less than all terms from the extended grammar. A second transcription of the utterance produced using the extended speech recognizer is obtained. The extended speech recognizer includes a speech recognizer that includes a language model trained by an extended speech recognition vocabulary including all terms of the extended grammar. The utterance is classified based on at least a portion of the first transcription or the second transcription.

Description

technical field [0001] The present disclosure generally relates to speech recognition. Background technique [0002] Speech recognition includes processing for converting spoken words into text. Typically, speech recognition systems map spoken utterances into a sequence of computer-readable utterances, and compare those utterances to known speech patterns associated with words. For example, a microphone may accept an analog signal that is converted into digital form that is then divided into smaller segments. Numeric fields can be compared to elements of the spoken language. Based on this comparison and an analysis of the environment in which those voices were emitted, the system is able to recognize the voices. [0003] A typical speech recognition system may include an acoustic model, a language model, and a dictionary. Briefly, an acoustic model includes digital representations of individual speech sounds that can be combined to generate a collection of words, phrases...

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): G10L15/32G10L15/197G10L15/30
CPCH04M2250/74G10L15/32G10L15/30G10L15/197G10L15/01G10L15/26
Inventor 彼塔尔·阿列克西克佩德罗·J·莫雷诺门希瓦尔法迪·比亚德希
Owner GOOGLE LLC
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