基于Java Speech API规范的语音识别引擎的实现  

Implementation of Speech Recognition Engine Based on Java Speech API Specification

作　　者：倪素萍[1] 董滨[1] 赵庆卫[1] 颜永红[1] 

机构地区：[1]中国科学院声学所中科信利语音实验室,北京100080

出　　处：《微计算机应用》2005年第2期168-172,共5页Microcomputer Applications

摘　　要：本文介绍了Java Speech API(JSAPI)规范的语音识别引擎的系统框架,描述了采用已有的C／C++识别引擎实现JSAPI语音识别引擎的思路和实现策略,提出并分析了以事件处理和状态处理为核心来实现JSAPI规范的具体方法,完成了基于JSAPI规范的语音识别软件系统的实现。In this paper, the framework of speech recognition engine defined by Java Speech API is introduced. The concept and strategy of implementing speech engine based on JSAPI using native recognition engine which is developed in C/C++ is described. The detailed implementation method focused on event and state processing is put forward and analyzed and speech recognition software system based on JSAPI is accomplished.

关 键 词：语音识别引擎 事件处理 JAVA SPEECH API规范 

分 类 号：TN912.34[电子电信—通信与信息系统]