[hts-users:02499] Re: AW: using of HVite

[Sébastien Le Maguer <Sebastien.Le_maguer@xxxxxxxx>]

That was it

thanks to all :)


Le vendredi 28 mai 2010 à 15:17 +0000, Florian Krebs a écrit :
> Hi Sebastien,
> 
> Did you also take into account that HVite does not understand HSMMs
> (Hidden Semi Markov Models), as they are generated by the HTS demo ?
> In order to use HVite you would have to turn the HSMM (monophone.mmf)
> into a HMM. This can be done as described by Simon King
> [hts-users:01141].
> You could also directly use a decoding method for HSMMs based on
> weighted finite state transducers (WFSTs), as proposed in
> [hts-users:01143] but I haven't tried this yet.
> 
> Good luck,
> 
> Florian Krebs
> 
> 
> 
> 
> 
> ______________________________________________________________________
> Von: Sébastien Le Maguer <Sebastien.Le_maguer@xxxxxxxx>
> An: hts-users@xxxxxxxxxxxxxxx
> Gesendet: Freitag, den 28. Mai 2010, 16:47:04 Uhr
> Betreff: [hts-users:02492] using of HVite
> 
> Hi,
> 
> I've got a problem with HVite. It always returns "No tokens survived
> to final node of network". I would like to find how monophones are
> aligned.
> 
> I used this command :
> 
> ./../tools/bin/HVite -A -B -C configs/trn.cnf -D -y lab -T 1  -o SW -f
> -a -H models/qst001/ver1/cmp/monophone.mmf -I data/labels/mono.mlf -i
> test/gen_align.mlf -S data/scp/train.scp data/dict/mono.dict
> data/lists/mono.list
> 
> mono.dict contains the same things that mono.list and all other files
> are generated from the HTS 2.1 demo (For the moment, I'm using HTS
> 2.1).
> 
> Here are samples of the messages I have:
> 
> Aligning File: <...>/demo/base/data/cmp/cmu_us_arctic_slt_b0538.cmp
> Created lattice with 31 nodes / 30 arcs from label file
> No tokens survived to final node of network
> Aligning File: <...>/demo/base/data/cmp/cmu_us_arctic_slt_b0539.cmp
> Created lattice with 32 nodes / 31 arcs from label file
> No tokens survived to final node of network
> 
> I think I'm misunderstanding something.
> 
> Thanks
> Sebastien
>