[hts-users:00710] Re: the score of HVite

["Heiga ZEN (Byung Ha CHUN)" <zen@xxxxxxxxxxxxxxx>]

Hi,

Heng Lu wrote (2007/06/14 10:23):

> When I use the tool HVite to calculate the score of the fulllab 
> of a sentence and the model used is the clustered context dependent model, 
> why the scores I get are all above 0 which is very different from the 
> log likelihood score in the speech recongnition (always a minus number) ?
> Is there anything wrong ? I use the 12 order mfcc parameter and 
> the mmf contains both the spectra model and the f0 model.

Are you using speaker-dependent model or speaker-independent model?

Regards,

Heiga ZEN (Byung Ha CHUN)

--
------------------------------------------------
Heiga ZEN     (in Japanese pronunciation)
Byung Ha CHUN (in Korean pronunciation)

Department of Computer Science and Engineering
Nagoya Institute of Technology
Gokiso-cho, Showa-ku, Nagoya 466-8555 Japan

http://www.sp.nitech.ac.jp/~zen
------------------------------------------------