语音识别_CMUSphinx入门(二)
目录
简介
上一篇学习了如何安装，并且体验了一下sphinx。但是最后还有个疑问就是为什么直接使用WAV识别是可以的，实时语音识别很差，所以接下来开始进一步了解sphinx。
学习步骤按照CMUSphinx的Tutorial(https://cmusphinx.github.io/wiki/tutorial/)进行，如下所示：
This tutorial is going to describe some applications of the CMUSphinx toolkit. Such applications could include voice control of your desktop, various automotive devices and intelligent houses. Other possible applications are speech transcription, closed captioning, speech translation, voice search and language learning. If you want to create one of them, the CMUSphinx toolkit is your choice.
这篇教程将会描述一些基于CMUSphinx工具包的应用。有些应用包括你的桌面语音控制、各种自动化设备以及智能家居，除此之外还可能是语音转录、内置字幕、语音翻译、语音搜索以及语音学习。如果你想创建的是其中一个，CMUSphinx工具包是你的选择。
The tutorial is intended for developers who need to apply speech technology in their applications, not for speech recognition researchers. If you are a researcher, it’s recommended to start with a textbook on speech technologies. Spoken Language Processing by Acero, Huang and others is a good choice for that.
这篇教程是写给那些准备在他们的应用中使用语音技术的开发者的，而不是语音识别研究者。如果你是一个研究者，这里推荐一本关于语音技术的书作为开始，Spoken Language Processing by Acero, Huang and others是一个好的选择。
The structure of this tutorial is the following:
- Basic concepts of speech recognition
- Overview of the CMUSphinx toolkit
- Before you start
- Building an application with sphinx4
- Building an application with pocketsphinx
- Using PocketSphinx on Android
- Building a dictionary
- Building a language model
- Adapting an existing acoustic model
- Training an acoustic model
- Tuning the performance
这篇教程的框架如下所示：
- 语音识别的基本概念
- CMUSphinx工具包概览
- 在你开始之前
- 基于sphinx4创建一个应用
- 基于pocketsphinx创建一个应用
- 在安卓上使用PocketSphinx
- 创建一个字典
- 创建一个语言模型
- 基于现有声学模型调整
- 训练一个声学模型
- 微调性能
语音识别的基本概念
参考https://blog.csdn.net/zouxy09/article/details/7941055，博主翻译的非常好。
CMUSphinx工具包概览
The CMUSphinx toolkit is a leading speech recognition toolkit with various tools used to build speech applications. CMUSphinx contains a number of packages for different tasks and applications. Sometimes, it’s confusing what to choose. To shed some light on the parts of the toolkit, here is a list:
- Pocketsphinx — lightweight recognizer library written in C.
- Sphinxbase — support library required by Pocketsphinx
- Sphinx4 — adjustable, modifiable recognizer written in Java
- Sphinxtrain — acoustic model training tools
We recommend that you use the latest available releases:
CMUSphinx工具包是一系列语音识别工具的总和，工具包里有各种工具用来创建语音应用。CMUSphinx包含多个部分，它们分别应用于不同的任务和应用。有时候，用户会困惑于该选择哪个。为了更好地说明工具包的各个部分，这里有一份列表：
接下来列出了当前最新的发布版本。
- Pocketsphinx — 用C语言编写的轻量级识别器库
- Sphinxbase — Pocketsphinx依赖的库
- Sphinx4 — 用Java编写的可调、可修改的识别器
- Sphinxtrain — 声学模型训练工具
- sphinxbase-5prealpha
- pocketsphinx-5prealpha
- sphinx4-5prealpha
- sphinxtrain-5prealpha
在你开始之前
基于sphinx4创建一个应用
……略
基于pocketsphinx创建一个应用
在安卓上使用PocketSphinx
……略