This is a script that allows some parts of the computer to be voice controlled. It is based on the mic_wav_streaming.py
of the DeepSpeech-examples, but expands some small features. Right now, only german inputs are supported.
When cloning, you need git lfs or you won't get the full repo!
python3 voicecontrol.py --model 51.565295.output_graph.pb --scorer de_kenlm.scorer --helpspeech
See install.sh
This is very much work in progress. Not all commands are listed (check source code for all commands). Also, the german language model is far from finished!
python3 voicecontrol.py --model 51.565295.output_graph.pb --scorer de_kenlm.scorer
Edit the file /usr/local/lib/python3.7/dist-package/wmctrl.py and a change line 2 to
from subprocess import getoutput