[Subject Prev][Subject Next][Thread Prev][Thread Next][Date Index][Thread Index]

[hts-users:00493] Re: HTK 2.0 hts_engine problem


Nicholas Volk a écrit :
When I comment the byte swapping line in misc.cpp
I get the problem I described in my earlier mail:

   //HTS_ByteSwap(p, size, block);

If the line is uncommented I get an error:
"Error: LoadModelFiles: #duration pdf must be positive value (-419364864)."

I think that the error is caused by endianism.
Thus I'd expect the commented version work fine, but the content
of the created raw is not really speech.

Hi Nicholas, once the raw is created how do you open it ?

because when I open it using wavesurfer, I have the choice between little and big endian. The good one is little endian.

I just tested the new hts_engine 2.0 with RMS and with HTS_ByteSwap(p, size, block); being commented and it works fine as long as I open the audio raw in little endian. Otherwise you get a kind of white noise.

Alexis


Follow-Ups
[hts-users:00494] Re: HTK 2.0 hts_engine problem, Nicholas Volk
References
[hts-users:00489] HTK 2.0 hts_engine problem, Nicholas Volk
[hts-users:00490] Re: HTK 2.0 hts_engine problem, Heiga ZEN (Byung Ha CHUN)
[hts-users:00491] Re: HTK 2.0 hts_engine problem, Nicholas Volk