[Accessibility] Re: [Ghmm-list] Can you help write a free version of HTK?

[Thomas Harris]

I've studied some of the HTK code, made some contributions to Sphinx, and I've modified and used ghmm for some other education-informatics research. Yes, recognizers like HTK employ hmms, but I would say that there's a world (say decades of research) of difference between having some unrelated hmm code and having a modern speech recognizer.

I have to agree with Peter on the angry-letter-to-everyone not being the best way to start this discussion in any case. I could be wrong, but I also tend to think like Peter that the read-and-rewrite-code tactic seems like a great way get sued.

If the HTK license doesn't suit your needs then you probably want to put hooks in to 
integrate Sphinx or something as an alternative. It doesn't sound like 
that would be too much work. It's never a bad idea to make your system 
more asr engine independent anyway. We always tried to make clean 
interfaces there when I was working on the Olympus spoken dialog 
architecture.

-Thomas

On Fri, Jul 9, 2010 at 2:52 PM, <address@hidden> wrote:

> On Jul 9, 2010, at 11:56 AM, <address@hidden> wrote:
>>> http://sourceforge.net/projects/ghmm/
>> I have already contacted them a while back and they said that their
>> library was not geared towards speech processing and couldn't be
>> used for
>> it neither now nor in the short term future IIRC.
>
> That is patently wrong.

Sorry, apparently I recalled it incorrectly. I lost my e-Mail archives a
while back so I couldn't check it again and to be honest I didn't look it
up in the online archives either. Sorry again.

I just remembered that GHMM doesn't have speech recognition as a goal
(which you stated again in your mail) and that at least back then it
couldn't replace the HTK completely (that's how I understood "I'm not sure
if it can be used for speech recognition" and "Most likely advanced HMMs
for speech recognition use some 'trick' which aren't implemented in the
GHMM. It won't be drop in replacement for HTK").

As I have no background in signal processing and thus wouldn't know the
first thing on how to get this working with ghmm, no matter how small the
inconsistencies are, this ruled it out as a solution for simon at that
time.

If the needed changes are fairly easy to implement, would you be
interested in working on this? Seeing as this would have a pretty big
impact on the open source speech recognition community we could maybe even
raise some funding for this...?

Regards,
Peter

> What Janne Grunau wrote to you on the GHMM-List 1/19/2009 was:
>
>> From: Janne Grunau <address@hidden>
>> Date: 21. Januar 2009 12:33:03 MEZ
>> On Monday 19 January 2009 12:01:13 Peter Grasch wrote:
>>> just wanted to know if GHMM can be used for that purpose.
>>
>> I'm not sure if it can be used for speech recognition. The basics HMM
>> building blocks are there but I don't know enough about the specific
>> HMM
>> usage in speech recognition. Most likely advanced HMMs for speech
>> recognition use some "tricks" which aren't implemented in the GHMM.
>> It won't
>> be drop in replacement for HTK.
>>
>>> Looking at your homepage and the sourceforge trove categorization, it
>>> seems that the generated models are not meant to be used for speech
>>> recognition. Does that pose a problem?
>>
>> Speech recognition is missing in the categorization since to my
>> knowledge
>> nobody uses the GHMM for speech recognition. As the name suggests
>> the GHMM
> ...
>
>>> Can the GHMM-generated models be
>>> used for that? (we use tied-state triphone models)
>>
>> I don't know what tied-state triphone models are exactly and short
>> internet
>> search wasn't conclusive.
>
> ...
>> specific features are likely to be missing. I can't compare it to
>> HTK since
>> even the HTK documentation requires a registration.
>
>
>
> AFAIK there was never a follow-up from you. Obviously, the GHMM is not
> a drop-in
> replacement for HTK, but there is is nothing which prevents you from
> using it. In fact
> there is a very good book by Rabiner about speech recognition laying
> out the details.
> Implement an HMM-based speech recognition system with the GHMM would be
> absolutely feasible; but as Janne pointed out we do bioinformatics   not
> speech
> recognition, so it is not done yet. Janne could have been more
> helpful, if at least
> the HTK docs would have been out in the open, or if you would have
> explained to him
> what the tied-state triphone models. In fact, there are just linear
> HMMs with the emission
> distributions tied. We do use tying for discrete emissions, extending
> to continous would
> be trivial.
>
> Best,
> Alexander
>
>
>
> --
> NEW Springer Book out: CATBox http://schliep.org/CATBox
> Dr. Alexander Schliep, Associate Professor,   address@hidden
> Department of Computer Science & BioMaPS Institute for Quantitative
> Biology
> Rutgers, The State University of New Jersey, Piscataway, NJ, USA
> Tel: (732) 445-2001 ext 1166   Hill Center #264




------------------------------------------------------------------------------
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
_______________________________________________
Ghmm-list mailing list
address@hidden
https://lists.sourceforge.net/lists/listinfo/ghmm-list