[hts-users:02408] Re: Any easy way of getting an adapted model set?

["Heiga ZEN (Byung Ha CHUN)" <heiga.zen@xxxxxxxxxxxxxxxxx>]

Hi,

Hui LIANG wrote:

> Your patch worked -- I obtained a non-diagonal covariance model set.
>
> By the way, I suppose the patch needs to be further improved. Currently, in any of the three following cases, an adapted model set will be generated by the AX command of HHEd:
>
> Case 1:
> HTK Configuration Parameters[26]
>   Module/Tool     Parameter                  Value
> # HADAPT          SAVEFULLC                   TRUE
>   HADAPT          SAVESPKRMODELS              TRUE
>   .... (other 24 parameters)
>
> Case 2:
> HTK Configuration Parameters[26]
>   Module/Tool     Parameter                  Value
> # HADAPT          SAVEFULLC                  FALSE
>   HADAPT          SAVESPKRMODELS              TRUE
>   .... (other 24 parameters)
>
> Case 3:
> HTK Configuration Parameters[26]
>   Module/Tool     Parameter                  Value
> # HADAPT          SAVEFULLC                  FALSE
>   HADAPT          SAVESPKRMODELS             FALSE
>   .... (other 24 parameters)
>
> This may be a problem especially when someone is going to generate an "adapted" HTS engine by both AX and CM commands. He may not want an adapted model set.

I knew this.  That patch code was for *you*, rather than for others.

Regards,

Heiga ZEN (Byung Ha CHUN)

--
--------------------------
Heiga ZEN (Byung Ha CHUN)
Speech Technology Group
Cambridge Research Lab
Toshiba Research Europe
phone: +44 1223 436975

______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
______________________________________________________________________