DSPRelated.com
Forums

Speech identyfication...

Started by Unknown February 15, 2006
Hi!!!
I have problem with speech identyfication. I use LPC algorithm and
extract reflection coeficient (rc). I have this 10 rc and try to
identify the person using it. The problem is how I can do this with
this rc??? I used Euklides distortion measures but I do not know what
can I do next with this :( Do You have any ideas how I can classificate
and identify the person using rc from LPC ???
thx for any help
Best Regards
Lukasz
PS> Sorry for my english....

malylulek@poczta.fm wrote:

> Hi!!! > I have problem with speech identyfication. I use LPC algorithm and > extract reflection coeficient (rc). I have this 10 rc and try to > identify the person using it. The problem is how I can do this with > this rc??? I used Euklides distortion measures but I do not know what > can I do next with this :( Do You have any ideas how I can classificate > and identify the person using rc from LPC ??? > thx for any help > Best Regards > Lukasz > PS> Sorry for my english.... >
Thou Shalt *NOT* apologize for your English :) [It is better than many native speakers] You might also wish to post to comp.speech.research . I suspect respondents on comp.dsp and comp.speech.research would come from instructively different points of view. I'm expert in NEITHER field but lurk on both. [It's very educational]
Thx
I will post to this group.
BS
Lukasz

malylulek@poczta.fm wrote:
> Hi!!! > I have problem with speech identyfication. I use LPC algorithm and > extract reflection coeficient (rc). I have this 10 rc and try to > identify the person using it. The problem is how I can do this with > this rc??? I used Euklides distortion measures but I do not know what > can I do next with this :( Do You have any ideas how I can classificate > and identify the person using rc from LPC ??? > thx for any help > Best Regards > Lukasz > PS> Sorry for my english....
Neural net techniques are often used for that sort of problem, and can work very well. Leon