Unlock instant, AI-driven research and patent intelligence for your innovation.

Pinyin-based voice wake-up method, device and computer equipment

A voice wake-up, computer program technology, applied in the computer field, can solve the problem of low sensitivity, achieve the effect of speeding up the response efficiency and avoiding adverse effects

Active Publication Date: 2020-08-11
深圳市友杰智新科技有限公司
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The main purpose of this application is to provide a voice wake-up method based on pinyin, aiming to solve the technical problem of low sensitivity of the existing voice wake-up

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
  • Pinyin-based voice wake-up method, device and computer equipment
  • Pinyin-based voice wake-up method, device and computer equipment
  • Pinyin-based voice wake-up method, device and computer equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0059] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application, and are not intended to limit the present application.

[0060] refer to figure 1 , the pinyin-based voice wake-up method of an embodiment of the present application, including:

[0061] S1: Receive the second probability vector obtained by calculating the second audio data by the specified neural network model, wherein the second audio data is the audio data corresponding to the second moment of the real-time audio stream;

[0062] S2: Output a specified probability value corresponding to the second audio data according to the second probability vector and the first previous probability vector updated at the first moment, wher...

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 application discloses a phonetic-alphabet-based voice awakening method. The method comprises the steps: receiving a second probability vector obtained by calculating second audio data by a specified neural network model; outputting a specified probability value corresponding to the second audio data according to the second probability vector and a first pre-posed probability vector updated atthe first moment; updating the first pre-posed probability vector according to a specified probability value corresponding to the second audio data to obtain a second pre-posed probability vector; outputting a specified probability value corresponding to the third audio data at the third moment according to the second pre-posed probability vector, and updating the second pre-posed probability vector until the real-time audio stream detection is completed; outputting a maximum probability value corresponding to a real-time audio stream and a phonetic alphabet sequence corresponding to the real-time audio stream; determining whether a target phonetic alphabet sequence corresponding to a preset wake-up word exists in the phonetic alphabet sequence or not; and if so, carrying out awakening according to the target phonetic alphabet sequence. With setting of the pre-posed probability vectors, real-time recording and updating of a detection result are realized, and the response efficiency isimproved.

Description

technical field [0001] The present application relates to the field of computers, in particular to a phonetic-based voice wake-up method, device and computer equipment. Background technique [0002] In the self-defined wake-up word system, or in the self-defined command word recognition model, the prior art generally adopts the method of phoneme sequence prediction or syllable classification, and compares the predicted phoneme sequence with the set phoneme sequence to determine the next step. operate. However, in speech recognition scenarios, complex situations such as homophones, near-phones, and polyphones must be faced, and efficient stream-based recognition cannot be achieved, resulting in poor wake-up sensitivity and poor wake-up effect. Contents of the invention [0003] The main purpose of the present application is to provide a voice wake-up method based on pinyin, aiming to solve the technical problem of low sensitivity of the existing voice wake-up. [0004] Th...

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): G10L17/24G10L17/04G10L17/18
CPCG10L17/04G10L17/18G10L17/24
Inventor 徐泓洋王广新杨汉丹
Owner 深圳市友杰智新科技有限公司