[Subject Prev][Subject Next][Thread Prev][Thread Next][Date Index][Thread Index]

[hts-users:03299] Re: How to improve performance of TTS without manual phoneme alignment


My opinion:
HTS system use phoneme boundaries only in the initialization stage of monophone models, which will affect the result of convergences in the embedded training of full-context models. But in my experiments, phoneme alignment does not affect that much. So I do not know what is your corpus size, maybe using more training data would help.
--
Xingyu Na (那兴宇)
Beijing Institute of Technology
naxy(at)bit.edu.cn
asr.naxingyu(at)gmail.com
naxingyu at {facebook, twitter, linkedin}

At 2012-05-13 03:29:51,"Kwan Lisa" <lisakwan1102@xxxxxxxxx> wrote: >Hi, > >I'm using a corpus without manual phoneme alignment. Thus, I performed >forced alignment to get the phoneme boundary information. >However, the performance of the TTS system was not good. TTS system >seems to be very sensitive to the accuracy of the phoneme boundary >information. >Is there any method that could improve the performance of TTS without >manual phoneme alignment? > >-- >Lisa Kwan >lisakwan1102(at)gmail.com >



Follow-Ups
[hts-users:03300] Re: How to improve performance of TTS without manual phoneme alignment, huanliangwang
References
[hts-users:03295] How to improve performance of TTS without manual phoneme alignment, Kwan Lisa