[hts-users:02450] Re: MSD-HSMM and WFST
- Subject: [hts-users:02450] Re: MSD-HSMM and WFST
- From: Keiichiro Oura <uratec@xxxxxxxxxxxxxxx>
- Date: Sat, 17 Apr 2010 18:23:31 +0900
- 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:sender:received:in-reply-to :references:date:x-google-sender-auth:received:message-id:subject :from:to:content-type:content-transfer-encoding; bh=w2kYWIQi0Bspn9YLxFa5/K8OpCfQkeiLGaXuldxrd1c=; b=faLTBdxeewUUXwQnzH5WvPYmOm1l9ZfsBfPM1n65hLNQK8+HP5INNzzByPCvyX2BjG M7ZJaA3p2G70tHrG69a6vitIcwJJMBmmb+xQFVbno6vUS1vLuOZUO1qSXSl4mAJI2Tzy /aDUmA/Zhl2pzVrnvcEtpfejhtGwN4Xt1xyWc=
- Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; b=IApVnFAIs4DaoJCCyeepcaQM0OjjaGdrvO1SmOmxqJoZXerWth+BtPIlpqifTuFOc2 v0Urbi70q64N+R7QXhuvaUIBzDu4mWI+huj/6iCdvcxoOWhJ39xykoLaMkH9Ms7X11Hf 9dBxIWytBBfObQcUtvmXhWA+5rKrZqWQdygsU=
Hi,
2010/4/17 Chongjia Ni <cjni@xxxxxxxxxxxxx>:
> Dear Keiichiro Oura:
> I have read your paper named "A fully consisten hidden semi-markov
> model-based speech recognition system". I want to ask you some questiones
> aboust hsmm and HTS.
> The first question is that the tool HFst in HTS 2.1.1 beta version can
> convert MSD-HSMM to FST?
Yes.
> The second question is HVite in HTS2.1 can support MSD-HSMM decode?
No.
> The last question is that how to use WFST to decode speech as your paper described?
Although I developed WFST decoder, I don't release it.
You can use WFST decoder such as 'Juicer'.
Regards,
Keiichiro Oura
>
> Thanks a lot!
>
> 2010-04-17
> ________________________________
> Chongjia Ni
- References
-
- [hts-users:02448] MSD-HSMM and WFST, Chongjia Ni