[hts-users:04538] Re: Preparing data for HTS
Hello,
what you need for HTS are :
- the questions
- the labels (full & mono)
- the audio
The utt are used as an input for festival to generate the
labels. Therefore, you do not need festival if you have another
way to generate the labels and the questions files.
To generate the questions and the labels files there is no
universal recipe it depends on your language. If it is a
phone-based language, you should at least you a tri-phoneme
context in order to capture coarticulation. For the rest of the
descriptive features it depends on what you can have and what is
important for your language.
Kind regards
Sébastien
On Sun, Jul 16 2017 (07:15), Atlas Khan <atlaskhan90@xxxxxxxxx>
wrote:
Hi,
I am working on Speech Synthesis for language which do not have
any type of
support in Festival. It has different phonemes and Lexicons than
English. I
have recordings in *raw *format. As per my knowledge, I need
following
types of data for speech synthesis with HTS.
1. questions
2. labels (full and mono)
3. utt
I want to ask how can I prepare questions, labels for language
which have
different lexicon and phonemes than English. If I need Festival
for
generating that data, than how can I do for language for which
Festival do
not have any support.
Regards,
Atlas Khan
--
Save our in-boxes! http://emailcharter.org
================================================================================
Dr. Sébastien Le Maguer
Postdoctorate researcher
Co-chair of SYNSIG (https://synsig.org/index.php/Main_Page)
Saarland University
Campus C7.4 - room 2.03
D-66123 Saarbrücken
Germany
phone : +49-681-302-70030
Mail: slemaguer@xxxxxxxxxxxxxxxxxxxx
website : http://www.coli.uni-saarland.de/~slemaguer/
================================================================================
- Follow-Ups
-
- [hts-users:04539] Re: Preparing data for HTS, Atlas Khan
- References
-
- [hts-users:04537] Preparing data for HTS, Atlas Khan