I am interested in developing a software that can recognise spoken tabla bols and will give the transcribed tabla notation as the output. .for this i require a large database of tabla bol compositions including the kaydas,paltas,tukras,etc. .can anybody help me get such material
plz mail ti saswat_nitj@rediffmail.com
That sounds like a serious mental gymnastic. Whether there is a point in it I don't know. All I know is that there is a ancient, older and simpler way of trancribing compositions. It takes only a pen and paper plus you are bound to get a nice hand writing and is easier to carry with you. Still, it would be interesting to see it.
To tell the truth, I would rather spend my time palying tabla.
cheers
Don't be so joyless , i think it is a good idea ,
even if it is not the work for a tabla player ,
I have a french friend who still did it , and it work
very good , just 5 percent of error .
It is a way to understand the way of playing of great master . I m ok to help you but i live in france so . .
Saswat,
I'd pretty much agree with Mewan. I don't wish to discourage anyone's inventive enterprise, but there's an inherent flaw in every attempt to transcribe the spoken word that no amount of engineering is likely to improve. That is the subtle inflexions and regional dialects that are sometimes critical to accurate interpretation of the meaning of a word or bol and the method that is employed to produce its counterpart. Sometimes I wish there was a universal method of transcribing tabla bols but then I also wonder how much of what cannot be transcribed would be lost? And, how much of what is lost is the something that we all seek but even the best find only rarely or late in the quest? I'm speaking of the music.
IMHO it might be interesting to see what a computer program would do with voice output of those same transcriptions, perhaps set to tempo and a background nagma?
Either way, go for it!
Aanaddha
