Detection method and device for false identification of similar sounds and computer equipment

A detection method and misrecognition technology, applied in speech recognition, speech analysis, instruments, etc., can solve the problem of high cost, achieve the effect of reducing the probability of misrecognition and good robustness

Active Publication Date: 2022-04-12
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 detection method, device and computer equipment for similar sound error recognition, aiming to solve the disadvantages of retraining the speech recognition model and high cost in order to solve the new similar sound error 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
  • Detection method and device for false identification of similar sounds and computer equipment
  • Detection method and device for false identification of similar sounds and computer equipment
  • Detection method and device for false identification of similar sounds and computer equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0019] 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.

[0020] refer to figure 1 , an embodiment of the present application provides a detection method for misrecognition of similar sounds, including:

[0021] S1: collect voice data, and input the voice data into the wake word phoneme recognition model for processing, and generate a decoding probability matrix;

[0022] S2: respectively determine whether the wake-up word path score of the wake-up word in the decoding probability matrix is ​​less than the first threshold, and / or whether the corresponding keyword phoneme probabilities of the keywords contained in the wake-up wor...

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 detection method and device for false identification of similar sounds and computer equipment, and the method comprises the steps: firstly, based on the size relation between a wake-up word path score and a first threshold value, achieving the most basic identification function for a correct wake-up word; furthermore, through the size relationship between the proportion of the keyword phoneme probability column corresponding to the keyword in the decoding probability matrix and a second threshold value, the targeted recognition of the keyword in the wake-up word is realized, and the probability of misrecognition is further reduced; furthermore, through comparison of the relative size relation between the wake-up word path score and the similar word path score, a fixed threshold value does not need to be set, the method is not influenced by far-field voice and near-field voice, the robustness is good, and when new similar voice appears, the method is easy to operate. The method only needs to compare the size relation between the wake-up word path score and the path score of the new similar sound during detection, does not need to re-train a wake-up word phoneme recognition model, is convenient and rapid, and is lower in cost.

Description

technical field [0001] The present application relates to the technical field of speech recognition, in particular to a detection method, device and computer equipment for similar-sound misrecognition. Background technique [0002] When the wake-up word and command word recognition technology is applied, it can monitor the user's voice in real time and provide corresponding feedback when specific words are detected (such as waking up the device when the wake-up word is detected). However, in practical applications, similar sounds and similar sounds are often misrecognized and falsely awakened, that is, when the pronunciation of the user's voice is close to the wake-up word, there is a certain probability that it will be misrecognized, thereby waking up the device by mistake; The wake-up situation will bring a very bad experience to the user. The existing processing for similar-sound false awakening usually adds negative samples to the training data to improve the ability of...

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
IPC IPC(8): G10L15/06G10L15/02
Inventor 李杰王广新杨汉丹
Owner 深圳市友杰智新科技有限公司
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

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