I am training acoustic model using HTK & checking phone level recognition at mono phone level & triphone level. The recognition rate at triphone level decreased 20-30% as compared to monophone recognition rate. Can anyone suggest possible reason for that.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I am training acoustic model using HTK & checking phone level recognition at mono phone level & triphone level. The recognition rate at triphone level decreased 20-30% as compared to monophone recognition rate. Can anyone suggest possible reason for that.
Narrow beams