[hts-users:00378] about HTS2.0RC's voice

["lei liu" <wenshiliu@xxxxxxxxx>]

Hi,

 

I have run the demo of HTS2.0's voice,

I find that the voice's  articulation is the same as HTS1.1 ,

 

even worse than that .

 

 

I find the f0s which in the HTS2.0's demo   have some prolems.

 

 

I use the Actival tcl and  getf0.tcl to generate .f0  for the cmu speech database that used in HTS1.1,

and compare them  with  the f0s that contained in HTS-demo_CMU-ARCTIC-AWB used in  HTS1.1,

 

I find there are some differences.

 

Does the snack' pitch  have some problems?

 

thank you!

Best regards