开源离线语音识别(SpeechRecognition)

我就是我 2022-08-31 05:58 222阅读 0赞

本次将使用SpeechRecognition和pocketsphinx进行离线语音识别。

**安装环境：**

操作系统 ： CentOS7.7.1908\_x64  
gcc版本 ：4.8.5  
Python 版本 : 3.6.8

> Centos7安装python3 参考：[https://blog.csdn.net/qq\_35385687/article/details/119026723][https_blog.csdn.net_qq_35385687_article_details_119026723]

**安装资料文件下载：**

链接: [https://pan.baidu.com/s/1Out0tJlb\_Qs-2C06\_2YHOQ][https_pan.baidu.com_s_1Out0tJlb_Qs-2C06_2YHOQ] 提取码: i6tg

**安装语音识别环境：**

pip install SpeechRecognition
    yum install python3-devel
    yum install pulseaudio-libs-devel
    yum install alsa-lib-devel
    pip install  PocketSphinx

**配置中文语音识别数据：**

下载地址：

[https://sourceforge.net/projects/cmusphinx/files/Acoustic%20and%20Language%20Models/][https_sourceforge.net_projects_cmusphinx_files_Acoustic_20and_20Language_20Models]

选择： Mandarin->cmusphinx-zh-cn-5.2.tar.gz

**安装中文语言包：**

cd /usr/local/python3.6.8/lib/python3.6/site-packages/speech_recognition/pocketsphinx-data
    tar zxvf cmusphinx-zh-cn-5.2.tar.gz
    mv cmusphinx-zh-cn-5.2 zh-cn
    cd zh-cn
    mv zh_cn.cd_cont_5000 acoustic-model
    mv zh_cn.lm.bin language-model.lm.bin
    mv zh_cn.dic pronounciation-dictionary.dict

**配置环境：**

cd /usr/local/python3.6.8/lib/python3.6/site-packages/speech_recognition/pocketsphinx-data
    tar zxvf py36asr.tar.gz
    source ./py36asr/bin/activate

**测试文本：**

自然语言理解和生成是一个多方面问题，我们对它可能也只是部分理解。

**语音识别示例：**

[root@localhost pocketsphinx-data]# pwd
    /usr/local/python3.6.8/lib/python3.6/site-packages/speech_recognition/pocketsphinx-data
    [root@localhost pocketsphinx-data]# ls
    cmusphinx-zh-cn-5.2.tar.gz  py36asr         test1.py   test2.wav  zh-cn.tar.gz
    en-US                       py36asr.tar.gz  test1.wav  zh-cn
    [root@localhost pocketsphinx-data]# python3 test1.py 
    自然 语言 李杰 和 申城 是一 个 多方 面 问题 我们 对 他 可能 也 只是 部分 礼节
    [root@localhost pocketsphinx-data]#

**test1.py**

# -*- coding: utf-8 -*-
    # /usr/bin/python
    
    import speech_recognition as sr
    r = sr.Recognizer()    
    test = sr.AudioFile("test1.wav") 
    with test as source:       
        audio = r.record(source)
    type(audio)
    c=r.recognize_sphinx(audio, language='zh-cn')    
    print(c)

**注意：**

*  出现speech\_recognition未安装问题，重新安装即可
 *  可根据需要调整为离线实时语音识别，参考项目为：[https://gitee.com/yzdyzdyzd/speechToText][https_gitee.com_yzdyzdyzd_speechToText]

[https_blog.csdn.net_qq_35385687_article_details_119026723]: https://blog.csdn.net/qq_35385687/article/details/119026723
[https_pan.baidu.com_s_1Out0tJlb_Qs-2C06_2YHOQ]: https://pan.baidu.com/s/1Out0tJlb_Qs-2C06_2YHOQ
[https_sourceforge.net_projects_cmusphinx_files_Acoustic_20and_20Language_20Models]: https://sourceforge.net/projects/cmusphinx/files/Acoustic%20and%20Language%20Models/
[https_gitee.com_yzdyzdyzd_speechToText]: https://gitee.com/yzdyzdyzd/speechToText