Technical discussions about the implementation and research of speech recognition algorithms.
Post a new Thread
rain into - May 14 2009
=C2=A0=C2=A0 i=C2=A0 want to do some recognition experiments on the TIMIT d=
atabase.=C2=A0 in the first place, i have to define the task grammar.
problem is what k...
Juan Macedo - Apr 8 2009
Hello everybody, I am working in a very interesting new approach to voice
recognition, in order to do that, the first step that I need is a very good
approach to vowel recognitio...
eng_maysa_noby - Apr 7 2009
iam working on a project of speech recognition for isolated arabic words,can any ony provide me with matlab neual network code for the recognition step.
Hasnat - Feb 23 2009
I want to develop a speech recognizer for my language. I want an
advice about selecting the best tool for the development of an speech
recognizer. I heard ...
hyee...@yahoo.com.cn - Feb 23 2009
1. As a kind of cepstrum, mfcc must comply with the basic process of cepstrum computation.
But while calculating mfcc,Why use "dct",not "idft" ?
Here,I do not deny the rat...
HMM [2 Articles]
ven_...@hotmail.com - Feb 23 2009
I am also looking for a HMM implementation in C. Any luck with your search for an Open Source?
ramy...@gmail.com - Jan 8 2009
Iam doing my final year project based on "Speaker Identification using instantaneous frequencies" and we are planning to implement it in matlab environment. But we are currently he...
jain...@yahoo.com - Oct 8 2008
I am working on a project on Speech enchancement. Can anyone provide me with MATLAB codes for Speech End point detection( using various parameters and methods). Its impor...
sachin joshi - Sep 18 2008
I am using sphinx2. I have a problem with 'sphinx2-simple' script. When I ran it, it gives quite
accurate output for first utterance. While from second utterace
onwards, the perf...
ns_nehe - Sep 11 2008
Hi all, can anybody provide me MATLAB code for 1) Gaussian distribution
with zero mean and unity variance random number generator and 2)
Exponential distributed, positive with u...