speech recognition algorithm

can anyone explain to me this web to me ??h**p://www.eecg.toronto.edu/~aamodt/ece341/speech-recognition/
what is the mean of the dictionary? a register? how to store the memory to dictionary?

anyone know? please help me, please...

I'm not shure what you are asking what is "mean of the dictionary" or what mean "dictionary" ?
If it is the first you should elaborate more to catch what you are asking.
"Dictionary" is used for any ASR system. btw. I think the correct therm is "Vocabulary". If we say that your system should recognize only 3 commands "one", "two" and "three" then your Vocabulary/Dictionary is {one, two, three}. Do not think as a register, thik like it is a very complex array in the memory. In Matlab that could be a "cell" where:
Vocabulary{1,1}='one';
Vocabulary{1,2}=fingerprint_of_one;
Vocabulary{2,1}='two';
.... etc
btw. the therm "fingerprint" is also something that I see for first time, normally is used "features".
If there is something else, just ask, but be more precise in your question.

can explain more detail, i'm still don't understand.
cell is like a matrix?
can u give to some guide step to do this ?
i'm still don't understand the "create the fingerprint and store into the memory dictionary".
sorry my english is weak.

cell is like matrix , but each element could be from a different type, even a structure. If you see the example I gave the elements in column 1 is a string, while elements in column 2 could be a floating value for example (if fingerprint_of_one=2.3).

Let assume you calculate a short time energy of a speech signal (I suppose you know at least this one). You divide the recording on, let say 3 segments, then the vector with short time energies for "one" should look like:
fingerprint_of_one=[ 5.7 4.8 2.1];
doing the same for the word "two" colud result into some other valuse, for example:
fingerprint_of_two=[ 2.3 3.5 8.8];
then for an unknown word ([3.1 3.8 7.1]) you have to calculate the distance measure (euclidean in your case) to the two fingerpints and to check which one is the closest match.
Now, if you expand the example from my previouse post with todays vectors, type them in Matlab and press enter, they will be stored in Matlab's working memory under the variable "Vocabulary". They will stay there until you close Matlab. Therefore after the initialisation you may recognize any word by comparing its distance to the different fingerprints (Vocabulary{:,2})

Is it clear now? Hope you can get it, but actually you have to read a lot of books/papers to do such a system.

i still blur, nvm i try the step waht u mention before..thank you, if i still dun understand can ask u again?

Ofcource, but remember. Asking the right question is an art and means you already answered 50% of it

hello, drago i'm still cant find the fingerprint of my word (voice). after i recognize a word, i only can see the data is only a array editor file, not vector file.

Finding the fingerprint is actually the "recognition". If you have recognized the word this means that you have calculated its fingerprint and then you have found the closest match to the database (fingerprints in the memory).
The fingerprint is "calculated" (not found), for example:
fingerprint_one=finerprint_function(time_domain_wav_file);
You can se the fingerprint (its values) by typing the variable and then press enter (without ";").

dargo can guide me some step to do it? i really dun have idea to do this , i'm sorry of disturbing u.

Let us do the following. Send to me your source (m files) to see what you are doing!
See your personal message, there is my email.

i have send the m file to u by mail ady

huh????

can any one plz send me the tor aamodt's algorithm