[hts-users:00161] Re: CMU-ARCTIC-AWB speech

["Heiga ZEN (Byung Ha CHUN)" <zen@xxxxxxxxxxxxxxxx>]

Hi,

jiang yuan wrote:

> I have Installed HTS-demo_CMU-ARCTIC-AWB , and try to synthesize
> speech by HTS_1.1.1. but it sound less nature and cadenced than you demo in
> the web. Because I had found many errors in the festival utterances for
> CMU-ARCTIC-AWB, I want ask you :
>
> 1.did you use the festival utterances or improved labeling?
> 2.Is the questions for lab in HTS-demo_CMU-ARCTIC-AWB same to the web demo system used?

Completely the same data, labels, scripts, questions were used.

> 3.Is there any other way to improve my synthetic speech?

In demo on festival website, postfiltering-based formant emphasis is performed.
It improves the naturalness of synthetic speech very much.

You can perform postfiltering using hts_engine (-b option).
There is a thread which discussed about hts_engine and postfiltering.
Please see

	http://kt-lab.ics.nitech.ac.jp/hts-users/spool/2004/msg00002.html

for detail.

Regards,

Heiga Zen

--
 ------------------------------------------------
  Heiga ZEN     (in Japanese pronunciation)
  Byung-Ha CHUN (in Korean pronunciation)

  Department of Computer Science and Engineering
  Graduate School of Engineering
  Nagoya Institute of Technology
  Japan

  e-mail: zen@xxxxxxxxxxxxxxxx
     web: http://kt-lab.ics.nitech.ac.jp/~zen
 ------------------------------------------------