基于隐马尔可夫模型的语音识别(语音模型)  

Speech Recognition Based on Hidden Markov Model(Speech Model)

作　　者：陈文云 Chen Wen-yun(Baise University,Baise 533000,Guangxi Zhuang Autonomous Region,China)

机构地区：[1]百色学院,广西百色533000

出　　处：《科学与信息化》2022年第21期37-39,共3页Technology and Information

摘　　要：语音识别是人工智能的重要分支,是实现智能化人机接口的重要方式。本文深入阐述了实现语音识别的主要方法,着重介绍了其中最为流行的基于隐马尔可夫模型的语音识别的基本原理和理论依据。它具有智能、准确寻找音节边界、语音识别率高的特点,在各种语音人机接口应用中被广泛应用。Speech recognition is the important branch of artificial intelligence and works as an important way to achieve intelligent human-machine interface.This paper deeply explains the main methods to achieve speech recognition and deliberately introduces the basic principle and theoretical evidence of speech recognition based on hidden Markov model,which is the most common model.This model has characteristics of identifying syllable boundary intelligently and accurately and high speech recognition precession,and has been widely used in various speech sound human-machine interface.

关 键 词：语音识别 自然语言处理 语音模型 语言模型 隐马尔可夫模型 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]