[hts-users:01474] Re: Trying new language on HTS

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

Hi,

Anocha Rugchatjaroen wrote (2008/06/25 17:15):

> Sorry, I forgot to tell you about our sampling rate. We record the speech
> in 44k sampling rate and resampling to 16k and train to the system.

Are you using 16kHz sampling and 5ms frame shift?
If so, your hts_engine command line is strange, it should be 

hts_engine -td tree-dur.inf -tf tree-lf0.inf -tm tree-mcp.inf -md dur.pdf \
          -mf lf0.pdf -mm mcp.pdf -dm mcp.win1 -dm mcp.win2 -dm mcp.win3 \
          -df lf0.win1 -df lf0.win2 -df lf0.win3 -or tmp.raw \
          -s 16000 -p 80 -a 0.42 -b 0.4 tmp.lab

Please also check your data/Makefile whether your mel-cepstral analysis 
condition is proper or not.

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