[hts-users:02896] About wav files recorded and utterances generated.
- Subject: [hts-users:02896] About wav files recorded and utterances generated.
- From: kulesh kumar <kulesh222@xxxxxxxxx>
- Date: Wed, 15 Jun 2011 22:52:08 +0530
- Delivered-to: hts-users@xxxxxxxxxxxxxxx
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:date:message-id:subject:from:to :content-type; bh=j8o2jYDRvnLodrs6BrbUYVdwMKw1WePc2tyQF93X9QM=; b=lSmGk2KhchFBKa72vugAOSgDGVjpTcG6+E1K4/GkuX0W3kReNI0NTHhHQVcnIFe1Sf U59LZlXEUkT6nB+9fScfFjRIm1YWvUwEABou0dpyG3Q+f9r9aVFRQqQJEcDXXNcCBHKv 8Kf6b1MdkjPhs6MXeY2e+vL05R/szri0J81O0=
- Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; b=lxeddQs3UiaT3zm77cvv4DdYpdwVewSL0rzda4v3gu5Bj25MQiGPocUuuCWWtkXoid 6jsC+oNLlXrLdLyiSLG9qaIx57WTvARVF6QO4PWfICEYr9GalCihvOziYnmW+NSUsr8+ q5oAjBZNKWdXFcfJu6KByTRRIllj8eLSqAR7o=
I have manually recorded voice for training process in hts for hindi language .Since Festival also supports my language , I am getting utterance file using festival.But during utterance
generation , you can hear the corresponding voice of festival , which is different from manually recorded files. I am afraid that this is creating problem because the duration factor and
stress factor might be different for the recorded one and generated one . This will in turn effect the label generation. I have completed the training process for my language .But the
is that the generated voice in gen folder sounds only as a noise.PDF are all ready but the speech synthesis is only noisy. I am attaching one sample wav file and the corresponding
utterance file.
please tell me if I am wrong.
one thing like in HTS demo inside data folder lab format is provided,Is the same applicable for Hindi language.
Thanks
Kulesh
Attachment:
exp01.utt
Description: Binary data
Attachment:
exp01.wav
Description: Wave audio