Dear Fang
my error 2021 due to the absence of the phonemes boundaries as doctor Heiga told me
i found that can escape from this step and go forward to HERest
Doctor Heiga adviced me to determine the phonemes boundaries if i have the a speaker-independent for my language then run HVite with -a option for the SI model
and due that u had see the same error that appeared to me i asked u
can u tell me the steps of the generation of SI model coz i see in ur question to doctor heiga that u have the boundaries for the phoneme a for example
Plz Help
i don't want to escape any steps in the HTS demo in order to generate a good speech quality
Thanks Alot Mr Fang Li
From: "Heiga ZEN (Byung Ha CHUN)" <zen@xxxxxxxxxxxxxxx>
Reply-To: hts-users@xxxxxxxxxxxxxxx
To: hts-users@xxxxxxxxxxxxxxx
Subject: [hts-users:00601] Re: MLSA filter
Date: Wed, 04 Apr 2007 21:52:38 +0900
>Hi,
>
>Kishore Prahallad wrote:
>
>>To use MLSA (mlsa_resythesis), is it necessary to use MCEP produced
>>from SPTK only?
>>
>>My question is: I have a feature extraction routine (NOT SPTK)
>>which generates MCEPs. Would it affect if I use these MCEPs to
>>resynthesize using MLSA?
>
>It works if these MCEPs are properly generated.
>
>Regards,
>
>Heiga ZEN (Byung Ha CHUN)
>
>--
>------------------------------------------------
>Heiga ZEN (in Japanese pronunciation)
>Byung Ha CHUN (in Korean pronunciation)
>
>Department of Computer Science and Engineering
>Nagoya Institute of Technology
>Gokiso-cho, Showa-ku, Nagoya 466-8555 Japan
>
>http://www.sp.nitech.ac.jp/~zen
>------------------------------------------------
>