[hts-users:00712] Re: the score of HVite

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

Hi,

Heng Lu wrote (2007/06/14 10:50):

> it is speaker-dependent
>
> > > 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) ?

Usually speaker-dependent model (trained using a few hours of speech) gives much higher likelihood to the training data than 
speaker-independent model used in speech recognition (trained using hundreds/thousands of hours of speech).

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
------------------------------------------------