[hts-users:00080] f0 extraction using pda on raw sound files

[Anders Lundgren <anderslu@xxxxxxxxxxxxx>]

Hi all,

I tried to extract f0 contours using the pda utility from the Edinburgh 
Speech Tools, but I'm not sure which command line parameters that work 
best with HTS. For instance, what low/high freq boundaries should be set 
when extracting a male voice? Also, are there any other parameters (ie 
voiced/voiceless treshold) that could affect HTS performance?

I created som f0 files using this utility and packed them to a binary 
float little endian, but I receive the "ViterbiAlign: No path found in 
8'th segment" when training reaches "sil" (silence). I have successfully 
trained using the exact same data, but with f0 contours taken from the 
KTH "Snack" f0 extraction tool. The problem then is that many segments 
in sentence-final position becomes partially unvoiced though there is no 
evidence for this in the training data.

Am I right in believing this might be a treshold problem, for the 
voiced/unvoiced parts of speech? Or, could it be that some frames have 
been truncated from the f0 files created with KTH Snack?