Dear All
I have some problems about HTS software,
Can someone give me some advice? the questions list is
below.
1. I want to synthesize new sentence using the English Speaker
dependent training demo,but i don't know how to do? So can you give me some
advice?
2. When I try to do some research about the HMM method training,I
find it is difficult to understand phoneme training. I wonder what the
state stand for?
Because the phoneme is
the minimum unit, so the different state stand for what
for the single phoneme?
3. For the Triphone HMM method training, what is the
observed quantity? and what is the state?
Does the different state stand for different the
phoneme? and the observed quantity is the parameter f0 or the
mell cepstrum, is this right?
4. I know the hts has no independent text analyser,it
can be done by the festival software. So I want to know what
is the format of the text analyser
output? Is it the ssml 1.1?
Thank you for your help in advance! 2010-12-02
bin007.zhao
|