Install GStreamer with pocketsphinx#
pocketsphinx within speech recognition.
Install gstreamer with pocketsphinx (Ubuntu 20.04 Focal)#
sudo apt install pocketsphinx libgstreamer1.0-dev libgstreamer-plugins-base1.0-dev gstreamer1.0-pocketsphinx
Next, you can verify the detection of Spanish sentences on the dictionary and models hosted in our speech repository.
cd ~/repos/speech/share/speechRecognition pocketsphinx_continuous -inmic yes -lm dictionary/20k-es.lm -dict dictionary/20k-es.dic -hmm model/es/
Check out the following sections if something goes wrong.
sudo apt install libgstreamer1.0-dev sudo apt install libgstreamer-plugins-base1.0-dev
Install pocketsphinx dependencies#
pocketsphinx from source,
sphinxbase is a dependency.
sudo apt install autoconf libtool cd # go home mkdir -p repos; cd repos # create $HOME/repos if it doesn't exist; then, enter it git clone https://github.com/cmusphinx/pocketsphinx cd pocketsphinx ./autogen.sh ./configure make -j$(nproc) # compile sudo make install # install
Configure gstreamer with pocketsphinx#
We need to configure the path to look for shared libaries:
echo "export GST_PLUGIN_PATH=$GST_PLUGIN_PATH:/usr/local/lib/gstreamer-1.0" >> ~/.bashrc
Verify gstreamer with pocketsphinx#
To verify that GStreamer can find the plugin. You should get a large amount of output, ending with something like this:
gst-inspect-1.0 pocketsphinx decoder : The underlying decoder flags: readable Boxed pointer of type "PSDecoder"
To test that all is running ok:
pocketsphinx_continuous -inmic yes
Troubleshooting GStreamer with pocketsphinx#
Most importantly, follow steps in correct order: first gstreamer, then pocketsphinx, and finally configure. If you install
gstreamer, you'll end up having to install
gstreamer first, then re-
pocketsphinx, compile and install again.
Some tutorials also recommend:
echo "export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/lib" >> ~/.bashrc echo "export PKG_CONFIG_PATH=$PKG_CONFIG_PATH:/usr/local/lib/pkgconfig" >> ~/.bashrc
As well as:
sudo modprobe snd_pcm_oss # Depends on your sound system