Hi,
I have trying Sphinx2-simple and the turtle demo gives about 50% accuracy for me. When the tty-continuous is running I would like to see where & how the phonemes are being generated by sphinx2 system for the current spoken word. Could anyone tell me as to how this can be done ?
Thanks
Edison
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
I have trying Sphinx2-simple and the turtle demo gives about 50% accuracy for me. When the tty-continuous is running I would like to see where & how the phonemes are being generated by sphinx2 system for the current spoken word. Could anyone tell me as to how this can be done ?
Thanks
Edison
Hi!
Are you American? I'm French and I have the same problem (sometimes, the recognition rate is below 30%... I have a bad English accent :-) )
I am also interested in finding how to display the phonemes to know where my accent is bad...
Julie