基于Android的语音识别系统

Speech Recognition System based on Android

笔者以Android手机开发平台为基础,通过分析Pocket Sphinx开源语音识别引擎的工作原理,提出了基于Android的离线语音识别系统。介绍了Android手机开发平台的系统架构和JNI、NDK等技术,并将Android系统移植到OMAP35x平台。最后在基于Android的智能手机上实现了离线语音识别系统,和过去使用的网络在线语言识别系统相比,该系统的最主要优点就是成本较低、应用场合广泛,可以应用于本地语音输入法、智能玩具等。

This paper is based on the Android mobile phone development platform,through the analysis of working principle of PocketSphinx open source speech recognition engine,proposes off-line speech recognition system based on Android.This paper introduces the system architecture and JNI Android mobile phone development platform,NDK technology,Android system is ported to the OMAP35x platform.The off-line speech recognition system is realized in the end intelligent mobile phone Android,online speech recognition system this system compared with the traditional low cost,wide application,can be applied to the local speech input method,intelligent toy,etc.