[hts-users:00479] 回复: Re: two questions
- Subject: [hts-users:00479] 回复: Re: two questions
- From: Fang Li <lifang3cis@xxxxxxxxxxxx>
- Date: Tue, 19 Dec 2006 09:04:33 +0800 (CST)
- Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com.cn; h=X-YMail-OSG:Received:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID; b=cIYYVvAfuO6ZSL1oUPRZOwb50XB9jg3Mc3KF7hKSuY1IRiF/CRZb+T5/6OWhpXNukR35/Ao2Shy+rQ7mD7ihLHM/G31Ph3bIWatmZWNj/aUoLYylaWnEdmrYuI5SzT6vGTrtcScYy1NhgmmU5prdFh49YQZnGWNITlN/aasftfU=;
Hi Alexis,
thanks for your explanation, you made the two conceptions easy to understand.
regards
Alexis Moinet <alexis.moinet@xxxxxxxxxxxxxxx> 写道:
Fang Li a écrit :
> hi,all
> I have two questions about duration model.
> First, I feel puzzled about two conceptions: the phone duration and
> state duration of HMM
> Do they refer to the same thing?
Not really, one phone is a sequence of n states (in the HTS demo, I
think it's a succession of 5 states). Therefore, the duration of 1 phone
is the sum of durations of these n states
Alexis
__________________________________________________
赶快注册雅虎超大容量免费邮箱?
http://cn.mail.yahoo.com
- Follow-Ups
-
- [hts-users:00480] error about HMGenS, Fang Li
- References
-
- [hts-users:00478] Re: two questions, Alexis Moinet