[hts-users:03393] Re: Is it possible to implement a HMM-based TTS system without MSDHMM?
- Subject: [hts-users:03393] Re: Is it possible to implement a HMM-based TTS system without MSDHMM?
- From: "Heiga ZEN (Byung Ha CHUN)" <heigazen@xxxxxxxxxx>
- Date: Thu, 9 Aug 2012 16:16:33 +0100
- Delivered-to: hts-users@xxxxxxxxxxxxxxx
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type:x-system-of-record; bh=9fwR1yFOjhhuSGDVVz5C+eb+/tmXDl5x/iayiS/g7Vo=; b=GFBeHPOs/9xO/6KdvfgM6BFTbNGYsu+rHLY7CAyrbeTBEidYG4WkzNO7oR9JhrZsRc FWkaLtVmI8HQGjo3oZHssljD7Uk9YeWTgWdMa3go2bxu66vJkI5qWyyk2GhANaDVhi8m pTD8x0gJ0TFxS9609q67MzxUA3JPxs8zfSwMWoHLN++D+0FCDOTi9VTu5xM/uMx5LqAX 92vip+bpSPAhRSn8cKZ6tOg7pbnmi5h4zuj3viVrDOhc3dpzDYZrFA+57tMCAYwq8HLd fuplOrilNywpPG8PMO725TBQ5j/hL711pyN/DMbOHIipdTsKs78QAle6M/6Pc0264hVx DAlA==
There are some approaches to model F0 values without using MSD-HMMs, e.g.,
http://mi.eng.cam.ac.uk/~sjy/papers/yuyo11.pdf
http://dx.doi.org/10.1109/ICASSP.2011.5947410
Regards,
Heiga
2012/7/19 Kwan Lisa <lisakwan1102@xxxxxxxxx>:
> Hi,
>
> Is it possible to implement a TTS system without MSDHMM? In HTS,
> frequency is modeled using MSD-HMM(Multi-Space Distribution HMM). It
> seems to be essential for a HMM-based TTS system to use MSD-HMM to
> deal with unvoiced and voiced parts. How does a HMM-based TTS system
> model frequency before MSD-HMM was invented?
>
> --
> Lisa Kwan
> lisakwan1102(at)gmail.com
> Advanced Speech Technology Lab, ASTL
>
--
Heiga ZEN (in Japanese)
Byung Ha CHUN (in Korean)
<heigazen@xxxxxxxxxx>
- References
-
- [hts-users:03384] Is it possible to implement a HMM-based TTS system without MSDHMM?, Kwan Lisa