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

Decoding network system, voice recognition method, device, equipment and medium

A speech recognition and decoding network technology, applied in the field of information processing, can solve the problems of low keyword recognition efficiency, bloated and inefficient keyword decoding, and low accuracy, and achieves reduction of information capacity and memory occupation, targeted optimization, The effect of improving accuracy

Pending Publication Date: 2022-08-05
时擎智能科技(上海)有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] A weighted finite state machine (weighted finaite-state transducer, WFST) is usually used as a decoder for speech recognition, but this method is bloated and inefficient for keyword decoding
At present, the scale of the huge decoding network will be reduced to facilitate keyword decoding, but it will bring a series of negative effects, such as low efficiency and low accuracy of keyword recognition

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
  • Decoding network system, voice recognition method, device, equipment and medium
  • Decoding network system, voice recognition method, device, equipment and medium
  • Decoding network system, voice recognition method, device, equipment and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] Before introducing the embodiments of the present invention in detail, some terms in the embodiments of the present application are explained below to facilitate understanding by those skilled in the art.

[0025] 1. A phone is the smallest unit of speech divided according to the natural attributes of speech. It is analyzed according to the pronunciation action in the syllable, and an action constitutes a phoneme. Phonemes are divided into vowels and consonants.

[0026] 2. The weighted finite-state transducer (WFST) is used for large-scale speech recognition, and its state changes can be marked with input symbols and output symbols.

[0027] 3. A token is a data structure that records the score and information of a certain state at a certain moment in the decoding process. Starting from the initial state of the weighted finite state machine, the token is transferred along the edge with direction, and the change of the state during the transfer process can be reflected...

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 decoding network system, a speech recognition method, a speech recognition device, speech recognition equipment and a medium. The decoding network comprises a first direction network unit and a second direction network unit, the first directional network unit comprises N first keyword transmission channels, and phoneme sequences on the first keyword transmission channels comprise first phoneme sequences of keywords; the second direction network unit comprises M second keyword transmission channels, and phoneme sequences on the second keyword transmission channels comprise second phoneme sequences of the keywords; and the second zero in-degree node is at least connected with one first zero out-degree node. The decoding network system provided by the invention can be only used for identifying and matching the keywords in the voice, so that the information capacity and the occupied memory are reduced while the keyword identification efficiency is ensured, and the keyword identification efficiency in voice identification is improved because the decoding process does not need to be backtracked.

Description

technical field [0001] The present invention relates to the technical field of information processing, and in particular, to a decoding network system, a speech recognition method, an apparatus, a device and a medium. Background technique [0002] Speech recognition technology, also known as automatic speech recognition (asr), aims to convert lexical content in human speech into computer-readable input, including keystrokes, binary codes, or character sequences, so as to realize human-computer interaction. Speech recognition technology has a wide range of application scenarios in modern life, and can be used in car navigation, smart home, voice dialing, simultaneous interpretation and other scenarios. The decoder is the core of the speech recognition system, and the speech decoding process based on the decoder plays an important role in the whole speech recognition process, which directly affects the accuracy of the recognition results. [0003] Usually, weighted finite sta...

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): G10L15/08G10L15/02
CPCG10L15/08G10L15/02G10L2015/025
Inventor 周智鄢戈仇健乐于欣蒋寿美
Owner 时擎智能科技(上海)有限公司