Hello everybody,
at first a big thanks for the nice software and all the support !
I installed pocketsphinx on my windows computer (with englich model and dict) and its working but the recognition quality is very very very bad. After my research of the tutorial and this forum, i guess the poor quality is because the sampling rate (?) of pocketsphinx and my windows microphone. The microphone settings of Windows allow me only to choose between 16bit 44.1 kHz and 16bit 48kHz.
Can anybodey help me and advice me how i can improve the quality of the voice recording ?
Thanks !!!!!
DerDom
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello everybody,
at first a big thanks for the nice software and all the support !
I installed pocketsphinx on my windows computer (with englich model and dict) and its working but the recognition quality is very very very bad. After my research of the tutorial and this forum, i guess the poor quality is because the sampling rate (?) of pocketsphinx and my windows microphone. The microphone settings of Windows allow me only to choose between 16bit 44.1 kHz and 16bit 48kHz.
Can anybodey help me and advice me how i can improve the quality of the voice recording ?
Thanks !!!!!
DerDom
You can record the wav file and try to recognize it. Share the file if there are problems.