Echo elimination method for voice recognition scene

A technology of echo cancellation and speech recognition, which is applied in speech recognition, speech analysis, instruments, etc. It can solve the problems of not having a variety of audio, and achieve the effect of enhancing operation experience, accurate recognition, and reducing interference

Active Publication Date: 2017-01-04
ARCHERMIND TECH WUHAN
View PDF7 Cites 32 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] Therefore, the method for obtaining the reference audio signal used for echo cancellation in the existing implementation of echo cancellation, and the implementation of the implementation does not have the function of eliminating various audio played by the device itself (such as playing audio and video, voice broadcast, etc.), and is not suitable for Speech Recognition Scenarios

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
  • Echo elimination method for voice recognition scene
  • Echo elimination method for voice recognition scene

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] The technical scheme of the patent of the present invention will be described in detail below in conjunction with the accompanying drawings and examples of implementation.

[0026] refer to figure 2 To implement echo cancellation, two kinds of audio data need to be obtained, namely the input audio data mic_input recorded from the microphone and the audio data speaker_output output from the speaker, because the microphone input audio data mic_input is the recording of all external sounds, which includes some speakers Output the content of the audio data speaker_output. The echo cancellation mentioned here is to eliminate the content of the speaker output audio data speaker_output contained in the microphone input audio data mic_input, and then output a clear speech_input voice. The ultimate goal is to enable the device to recognize external voice commands , expressed in a simple mathematical formula as:

[0027] speech_input = mic_input – F(speaker_output).

[0028] I...

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 belongs to the field of voice recognition, and provides an echo elimination method for a voice recognition scene. According to the method, double digital microphone channels are used. Microphone input and loudspeaker output audio data are simultaneously acquired in an audio digital signal processing module, and the loudspeaker output right channel data are copied to the right channel of the microphone input audio data so as to form synthetic microphone input audio data. The synthetic microphone input audio data are provided to an upper echo elimination module, and algorithmic processing is performed on the left and right channels of the synthetic microphone input audio data through the echo elimination module AEC so that voice input audio data which can be used for a voice recognition module are outputted, and equipment is enabled to recognize external voice instructions. The interference of the audios played by the equipment on the voice recognition function of the equipment can be effectively reduced so that the equipment is enabled to more accurately recognize the external voice instructions, and the operation experience of human-computer voice interaction can be enhanced.

Description

technical field [0001] The invention belongs to the field of voice recognition, in particular to a method for echo cancellation in a voice recognition scene. Background technique [0002] With the development of artificial intelligence technology, voice, as a good human-computer interaction mode, has gradually been applied to many smart devices. Speech input is an input method that converts what people say into text through speech recognition. [0003] In many fields, users can use voice input instead of manual input to execute corresponding commands. For example, users can open the music player application by saying "please turn on the music" to the mobile phone, and then say "play the next one" during playback. "Song" to switch songs. After the mobile phone recognizes the voice signal input by the user, it performs corresponding operations. [0004] However, in many speech recognition usage scenarios, if the device used to recognize the input voice may be playing audio ...

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): G10L21/02G10L21/0208G10L15/22
CPCG10L21/02G10L21/0208H04L15/22
Inventor 赵成义何川
Owner ARCHERMIND TECH WUHAN
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