我是在 arm cubieboard 下做的，i386下应该一样。
对了，在实际做之前先看一下 http://www.chenwang.net/2013/11/21/cmu-sphinx-%e8%af%ad%e9%9f%b3%e8%af%86%e5%88%ab%e5%ad%a6%e4%b9%a0%e7%ac%94%e8%ae%b0%ef%bc%881%ef%bc%89/ 文章里面有中文语音识别模型，下面的那个例子使用了中文模型，需要下载下来和 03.py 放到一个目录。
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 | sudo apt-get install gstreamer0.10-alsa gstreamer0.10-tools libgstreamer0.10-0 libgstreamer0.10-dev python-gst0.10 python-gst0.10-dev libgstreamermm-0.10-2 libgstreamermm-0.10-2 make gcc g++ bison libbison-dev libasound2-dev libasound2-python python-alsaaudio python-pyalsa python-gtk2 soundconverter vim sudo aptitude install libgstreamer-plugins-base0.10-dev # 这个事debian上面的，应该一样的。 sudo apt-get install gstreamer0.10-alsa gstreamer0.10-tools libgstreamer0.10-0 libgstreamer0.10-dev python-gst0.10 python-gst0.10-dev libgstreamermm-0.10-2 libgstreamermm-0.10-2 sudo apt-get install make gcc g++ sudo apt-get install bison sudo apt-get install libbison-dev sudo apt-get install libasound2-dev libasound2-python python-alsaaudio python-pyalsa sudo apt-get install python-gtk2 sudo apt-get install soundconverter wget http://downloads.sourceforge.net/project/cmusphinx/pocketsphinx/0.8/pocketsphinx-0.8.tar.gz?r=http%3A%2F%2Fsourceforge.net%2Fprojects%2Fcmusphinx%2Ffiles%2Fpocketsphinx%2F0.8%2F&ts=1385189168&use_mirror=softlayer-dal mv pocketsphinx-0.8.tar.gz\?r\=http\:%2F%2Fsourceforge.net%2Fprojects%2Fcmusphinx%2Ffiles%2Fpocketsphinx%2F0.8%2F pocketsphinx-0.8.tar.gz wget http://downloads.sourceforge.net/project/cmusphinx/sphinxbase/0.8/sphinxbase-0.8.tar.gz?r=http%3A%2F%2Fsourceforge.net%2Fprojects%2Fcmusphinx%2Ffiles%2Fsphinxbase%2F0.8%2F&ts=1385189298&use_mirror=softlayer-dal mv sphinxbase-0.8.tar.gz\?r\=http\:%2F%2Fsourceforge.net%2Fprojects%2Fcmusphinx%2Ffiles%2Fsphinxbase%2F0.8%2F sphinxbase-0.8.tar.gz tar xzf sphinxbase-0.8.tar.gz cd sphinxbase-0.8 ./configure make sudo make install cd .. export LD_LIBRARY_PATH=/usr/local/lib export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig tar xzf pocketsphinx-0.8.tar.gz cd pocketsphinx-0.8 ./configure make sudo make install pocketsphinx_continuous echo "export GST_PLUGIN_PATH=/usr/local/lib/gstreamer-0.10/" >> ~/.bashrc echo "export LD_LIBRARY_PATH=/usr/local/lib/" >> ~/.bashrc export GST_PLUGIN_PATH=/usr/local/lib/gstreamer-0.10/ linaro@cubieboard:~/yysb$ cat 03.py import pygtk,gtk,gobject gobject.threads_init() import gst def result(asr, text, uttid): print text pipeline=gst.parse_launch('alsasrc device=1 ! audioconvert ! audioresample ! vader name=vad auto-threshold=true ! pocketsphinx name=asr ! appsink sync=false name=appsink') #pipeline=gst.parse_launch('alsasrc ! audioconvert ! audioresample ! vader name=vad auto-threshold=true ! pocketsphinx name=asr ! appsink sync=false name=appsink') #pipeline=gst.parse_launch('pulsesrc ! audioconvert ! audioresample ! vader name=vad auto-threshold=true ! pocketsphinx name=asr ! appsink sync=false name=appsink') #pipeline=gst.parse_launch('pulsesrc ! tee name=t ! queue ! audioconvert ! audioresample ! vader name=vad auto-threshold=true ! pocketsphinx name=asr ! appsink sync=false name=appsink t. queue ! audioconvert ! audioresample ! wavenc ! filesink location=o.wav') asr=pipeline.get_by_name('asr') asr.connect('result', result) #asr.set_property('hmm', 'zh_broadcastnews_ptm256_8000') #asr.set_property('lm', '1.lm') #asr.set_property('dict', '1.dic') asr.set_property('configured', True) pipeline.set_state(gst.STATE_PLAYING) gtk.main() linaro@cubieboard:~/yysb$ python 03.py |
好了，现在可以测试了。
识别准确率固定几条命令的话识别率还行，但是会把其他的不相干的话识别成命令。现在正在尝试本地+在线语音识别，但是板子的音源输入只允许单个程序独占很烦人，现在正在考虑解决办法。
配置声卡参考这里：http://www.chenwang.net/2013/11/21/cubieboard-%E5%AE%98%E6%96%B9-ubuntu-%E5%A3%B0%E9%9F%B3%E9%85%8D%E7%BD%AE/