Adaptive voice activity detection method

A voice activity detection and adaptive technology, used in speech analysis, speech recognition, instruments, etc.

Active Publication Date: 2017-11-07
成都启英泰伦科技有限公司
View PDF7 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The current difficulties mainly lie in: 1) When multiple features are activated, different feature selections are used to achieve different frequency selection characteristics and different feature selections that take into account both frequency domain and time domain characteristics

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
  • Adaptive voice activity detection method
  • Adaptive voice activity detection method
  • Adaptive voice activity detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] The present invention will be described in further detail below in conjunction with the examples and specific implementation methods, but this should not be interpreted as the scope of the above-mentioned subject of the present invention being limited to the following examples, and all technologies realized based on the content of the present invention belong to the present invention scope.

[0035] Such as figure 1 Shown, a kind of adaptive voice activity detection method, it comprises the following steps:

[0036] In step 1, the time-domain signal is divided, and according to the set frame time and frame shift time, a long section of the time-domain signal is divided into a frame-by-frame single-frame time-domain signal. In this embodiment, the frame time is 32 milliseconds, the frame shift time is 16 milliseconds;

[0037] Step 2: Perform windowing processing on the single-frame time-domain signal using a Hamming window, perform fast Fourier transform on the single...

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 the voice signal processing field, and especially relates to an adaptive voice activity detection method mixing three thresholds and based on time domain and frequency domain characteristic extractions; the method can control the equipment work states by detecting whether a voice system has active voices or not; when no active voice is detected, the equipment can be in a low power state or a suspended state, thus reducing equipment operation requirements and power consumption, and reducing the communication bandwidth for the voice communication equipment.

Description

technical field [0001] The invention relates to the field of voice signal processing, in particular to a hybrid three-threshold adaptive voice activity detection method based on time domain and frequency domain feature extraction. Background technique [0002] At present, speech recognition, as a commonly used human-computer interaction technology, has been widely used in various electronic products. In the field of speech processing, as the starting point of all speech signal processing, the settlement result of the speech activity detection technology directly determines whether the subsequent processing is carried out. At the same time, for the speech recognition system, the speech activity detection is to achieve low power consumption and improve the recognition rate very important step. The current difficulties mainly lie in: 1) When multiple features are activated, different features are selected to achieve different frequency selection characteristics and different f...

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): G10L15/22G10L25/45G10L25/78
CPCG10L15/22G10L25/45G10L25/78
Inventor 何云鹏张来高君效许兵
Owner 成都启英泰伦科技有限公司
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