Re mlf files & the alphadigits prototype tutorial

[robert b <http://dummy.us.eu.org/robert>]

--- Joe Picone - The Terminal Man <http://www.isip.msstate.edu/~picone> wrote:
> Date: Tue, 8 Oct 2002 23:08:12 -0500 (CDT)
> From: Joe Picone - The Terminal Man <http://www.isip.msstate.edu/~picone>
> To: http://dummy.us.eu.org/robert
> CC: http://www.isip002.isip.msstate.edu/~parihar
> Subject: [http://www.isip.msstate.edu/~parihar: [http://www.isip.msstate.edu/~MAILER-DAEMON: Returned mail: see
> transcript for details]]
> 
> From: Naveen Parihar <http://www.isip.msstate.edu/~parihar>
> To: http://dummy.us.eu.org/robert]
> CC: http://www.isip.msstate.edu/~picone
> In-reply-to: <http://www.isip006.isip.msstate.edu/~200210090052.g990qGD196> (message
> 	from Joe Picone - The Terminal Man on Tue, 8 Oct 2002 19:52:16 -0500
> 	(CDT))
> Subject: Re: [http://www.isip.msstate.edu/~picone: FW: mlf files & the alphadigits prototype tutorial]
> Reply-to: http://www.isip.msstate.edu/~parihar
> References:  <http://www.isip006.isip.msstate.edu/~200210090052.g990qGD196>
> 
> > So, is there a way to derive the fs_num_states and fs_ci_models files from, say, the
> alphadigits
> > trans files?
> 
> The monophones or fs_ci_models file is the file that contains all the
> monophones used for the system. This can be extracted from the lexicon
> (pronunciations) used for the task. For alpha-digits, we use 18
> monophones + sp + sil. Beware, the lexicon should cover all the words
> in the transcriptions.
> 
> The fs_num_states file describes the phonetic-model topology and can
> be created manually once, you have the above file.
> 
> -Naveen