Hi all,
I have been playing with fft lately and was wondering if anyone has some pointers on how to extract consonants from vowels in human speech. I suspect it is detection of noisy signal that generally populates a much wider band of frequencies approaching some sort of filtered white noise (depending on the consonant) but am not sure as to what is the best strategy in isolating these at least somewhat reliably.
Any ideas/pointers would be most appreciated.
Ivica Ico Bukvic, D.M.A. Composition, Music Technology Director, DISIS Interactive Sound & Intermedia Studio Director, L2Ork Linux Laptop Orchestra Assistant Co-Director, CCTAD CHCI, CS, and Art (by courtesy) Virginia Tech Dept. of Music - 0240 Blacksburg, VA 24061 (540) 231-6139 (540) 231-5034 (fax) ico@vt.edu http://www.music.vt.edu/faculty/bukvic/
I09.sheep.from.goats.pd ? gr, Tim
2010/12/16 Ivica Ico Bukvic ico@vt.edu
Hi all,
I have been playing with fft lately and was wondering if anyone has some pointers on how to extract consonants from vowels in human speech. I suspect it is detection of noisy signal that generally populates a much wider band of frequencies approaching some sort of filtered white noise (depending on the consonant) but am not sure as to what is the best strategy in isolating these at least somewhat reliably.
Any ideas/pointers would be most appreciated.
Ivica Ico Bukvic, D.M.A. Composition, Music Technology Director, DISIS Interactive Sound & Intermedia Studio Director, L2Ork Linux Laptop Orchestra Assistant Co-Director, CCTAD CHCI, CS, and Art (by courtesy) Virginia Tech Dept. of Music - 0240 Blacksburg, VA 24061 (540) 231-6139 (540) 231-5034 (fax) ico@vt.edu http://www.music.vt.edu/faculty/bukvic/
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
I recently heard a seminar by a researcher called Gopala Anumanchipalli[1], here in Portugal. He works around similar subjects (modelling and extracting pronunciation via signal processing - and in order to to that I think he extracts consonants and vowels), take a look at his homepage maybe there's something there.
[1] http://www.cs.cmu.edu/~gopalakr/
also, http://www.cs.cmu.edu/~gopalakr/publications/Gopal_Thesis_final.pdf
Best regards, Pedro p.s.: maybe a too long shot. but what the heck. :)
On Thu, Dec 16, 2010 at 7:12 PM, tim vets timvets@gmail.com wrote:
I09.sheep.from.goats.pd ? gr, Tim
2010/12/16 Ivica Ico Bukvic ico@vt.edu
Hi all,
I have been playing with fft lately and was wondering if anyone has some pointers on how to extract consonants from vowels in human speech. I suspect it is detection of noisy signal that generally populates a much wider band of frequencies approaching some sort of filtered white noise (depending on the consonant) but am not sure as to what is the best strategy in isolating these at least somewhat reliably.
Any ideas/pointers would be most appreciated.
Ivica Ico Bukvic, D.M.A. Composition, Music Technology Director, DISIS Interactive Sound & Intermedia Studio Director, L2Ork Linux Laptop Orchestra Assistant Co-Director, CCTAD CHCI, CS, and Art (by courtesy) Virginia Tech Dept. of Music - 0240 Blacksburg, VA 24061 (540) 231-6139 (540) 231-5034 (fax) ico@vt.edu http://www.music.vt.edu/faculty/bukvic/
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-
Ivica Ico Bukvic -
Pedro Lopes -
tim vets