A method for solving the unknown of sensitive words in a speech recognition application system

A speech recognition and application system technology, which is applied in character and pattern recognition, instruments, computer components, etc., can solve the problems of poor sharing of speech recognition modules, no sharing, and inability to know input nodes, etc., to achieve a good human-computer interaction experience Effect

Active Publication Date: 2018-03-23
上海和汇安全用品有限公司 +2
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] 2) Different applications work independently, and the sharing of speech recognition modules is poor
Moreover, each application program develops a set of speech recognition subsystems separately, and there is no sharing
Different applications cannot know each other's input nodes, especially the detailed input node list, so it is difficult to truly share the speech recognition module

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
  • A method for solving the unknown of sensitive words in a speech recognition application system
  • A method for solving the unknown of sensitive words in a speech recognition application system
  • A method for solving the unknown of sensitive words in a speech recognition application system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] Such as figure 1 , figure 2 as shown,

[0058] figure 1 The sensitive word list in , is flexibly configured by the application program on the intelligent system using system mobilization, which effectively solves the unknown problem of sensitive words. The A7 system sends an interrupt to the main system A15 according to the result of the speech recognition algorithm; the main system A15 interrupts the service program accordingly, and transfers the speech recognition result and the corresponding application program process id and other information to the core area of ​​the operating system memory of the main system A15; The operating system of the system A15 adds a new system process, regularly queries the memory core area, and notifies the corresponding application program process through inter-process communication; the application program process responds to the corresponding behavior according to the received message, that is, The mapping of messages to input nod...

Embodiment 2

[0061] Such as image 3 As shown, only one CPU is used in this embodiment. image 3 The five modules of A) B) C) D) E) are the changes to the operating system. F in the figure) "sensitive word list and affiliated speech recognition module" is a hardware module, which can process speech recognition independently of the CPU (or co-processing), and sends an interrupt signal in due course according to the algorithm result; the corresponding interrupt service program will The voice recognition result and the corresponding application process id and other information are transferred to the core area of ​​the operating system memory; the operating system adds a new system process, periodically queries the core area of ​​the memory, and notifies the corresponding application process by means of inter-process communication; The process, according to the received message, makes a corresponding behavior response, that is, completes the mapping of the message to the input node. Similarl...

Embodiment 3

[0064] Such as Figure 4 As shown, embodiment 3 does not need the hardware module of "sensitive vocabulary and attached speech recognition module" in embodiment 2, and only needs to modify and supplement the existing operating system. Figure 4 The five modules of A) B) C) D) E) are the changes to the operating system. Figure 4 The voice recognition processing process uses the recognition algorithm to process the input voice signal, and transfers the voice recognition result and the corresponding application program process id and other information to the core area of ​​the operating system memory; the operating system adds a new system process, and periodically queries the core area of ​​the memory, And the corresponding application process is notified by means of inter-process communication; the application process, according to the received message, makes a corresponding behavior response, that is, completes the mapping of the message to the input node. Similarly, for the...

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 present invention relates to the technical field of intelligent information processing, and specifically relates to a method for solving the unknownness of sensitive words in a speech recognition application system: the sensitive vocabulary is introduced into the recognition algorithm, which reduces the time and space resources required for speech recognition, The sensitive vocabulary is a subset of the entire collection of sensitive words; the application continuously updates the sensitive vocabulary according to the progress of human-computer interaction and the change of the application scene; the result of the speech recognition algorithm of the sensitive vocabulary is modified by the operating system The interaction mechanism notifies the corresponding application process. The invention has the advantages of: introducing the sensitive vocabulary into the recognition algorithm, which reduces the resources required for each frame of speech recognition; the application program continuously updates the sensitive vocabulary, maintaining a good human-computer interaction experience; the recognition result is notified to the corresponding The application process enables all applications to share speech recognition function modules without consuming time and resources.

Description

[technical field] [0001] The invention relates to the technical field of intelligent information processing, in particular to a method for solving the unknownness of sensitive words in a speech recognition application system. [Background technique] [0002] Various contemporary speech recognition technology methods are roughly similar, usually divided into front-end processing, acoustic pattern training, and pattern matching (decoder). The most common front-end processing is to do endpoint detection (detect energy, find the starting point of speech) and speech enhancement (denoise, increase signal-to-noise ratio). The decoder is one of the cores of the speech recognition system. Its task is to find the word string that can output the signal with the greatest probability according to the input signal, according to the acoustics, language model and dictionary. The model of the speech recognition system usually consists of two parts: an acoustic model and a language model, cor...

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): G06K9/20
Inventor 徐彬宸
Owner 上海和汇安全用品有限公司
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