[mary-users] MARYTTS new voice training (HMMVoiceMakeVoice)

[Karthik Krishnan]

Hello all

I am new to marytts (and not a speech scientist, but a programmer). I am trying to get a new voice imported with marytts. I got to a stage where it is failing with "HMMVoiceMakeVoice". All the previous steps in the documentation until this have succeeded. The final extracted error is as below
---
ERROR [+2121]  HInit: Too Few Observation Sequences [0]
FATAL ERROR - Terminating program /usr/local/HTS-2.2beta/bin/HInit
------------

It sounds like I have too few training wav files ? Can anyone help me if there are other ways to overcome this ? Also is there a way to check if the wav files I have are sufficient or if there are any guidelines on the number of files etc ?  

Thanks
Karthik
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.dfki.de/pipermail/mary-users/attachments/20130721/c7df1ee9/attachment.htm