excuseme for repeat this part of this meesage before.
HI
Thanks for all the anwsers, and I can tell you now more clearly (I hope) th systme that I need to make.
I need recognize words 3 words HOLA ADIOS GRACIAS
I need a recognition system with Sphinx 4, speaker independent ( continous model)
Im work in Linux (Red-Hat 9.2)
The words must be in spanish-mexican
Can you tell me what is the less amount of data for this system (number of utterances, number of words/utterance, minimal length for any utterance (min), etc...)?
I need a language model?
I need run all the scripts, or only 00 and 02?
What is the structure of a utterances for this system ?
Can be:
The forum support on this site (SourceForge) is archived so that others can find/read discussions that may have already occurred. Two things help greatly with this: meaningful subject lines and message threads.
With message threads, one will create a "followup" to a message that pertains to a similar subject. That is, when discussing a particular subject matter, one will tend to keep it to a single thread by posting followups to a message rather than starting a new thread. When someone else searches for a related discussion that may have already taken place, a contiguous message thread will help them to reduce their search time greatly. A single thread with multiple followups will also allow them to quickly peruse the discussion in the order it happened.
In addition, the notion of a "subject" is also very important. For example, when someone interested in acoustic model training searches these forums, they might be more likely to examine messages with a subject line such as "Amount of data for needed for acoustic models" rather than "Hello" or "Thanks Will".
If you could follow the "subject" and "followup" conventions more closely, it will greatly help others who use the archives of these forums as a means to gather more information.
Thanks!
Will
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi
excuseme for repeat this part of this meesage before.
HI
Thanks for all the anwsers, and I can tell you now more clearly (I hope) th systme that I need to make.
Can you tell me what is the less amount of data for this system (number of utterances, number of words/utterance, minimal length for any utterance (min), etc...)?
I need a language model?
I need run all the scripts, or only 00 and 02?
What is the structure of a utterances for this system ?
Can be:
HOLA ADIOS GRACIAS
HOLA ADIOS GRACIAS
HOLA ADIOS GRACIAS
I need al the basics files (phone list, dictyonary, fill dictionary, trancription file control file)?
Thanks a lot
Omar
Hi Omar:
The forum support on this site (SourceForge) is archived so that others can find/read discussions that may have already occurred. Two things help greatly with this: meaningful subject lines and message threads.
With message threads, one will create a "followup" to a message that pertains to a similar subject. That is, when discussing a particular subject matter, one will tend to keep it to a single thread by posting followups to a message rather than starting a new thread. When someone else searches for a related discussion that may have already taken place, a contiguous message thread will help them to reduce their search time greatly. A single thread with multiple followups will also allow them to quickly peruse the discussion in the order it happened.
In addition, the notion of a "subject" is also very important. For example, when someone interested in acoustic model training searches these forums, they might be more likely to examine messages with a subject line such as "Amount of data for needed for acoustic models" rather than "Hello" or "Thanks Will".
If you could follow the "subject" and "followup" conventions more closely, it will greatly help others who use the archives of these forums as a means to gather more information.
Thanks!
Will