0

I am asking this question as the preliminary to start working on one of my fun-project. I am thinking of coding this in python so I guess, I have to go with pyAudio (please suggest me if there are better libraries for this other than pyAudio). My goal is to create a program that can record somebody's voice. This is simple. However, how can I identify the voice from the user after I have recorded his voice in one/multiple sample? What distinguishes one voice from other and how can python be implemented to achieve this?

Jack_of_All_Trades
  • 10,942
  • 18
  • 58
  • 88
  • 2
    Some simple voice recognition programs use MFCCs - have a look at [wikipedia](http://en.wikipedia.org/wiki/Mel-frequency_cepstrum) and [this](http://yaafe.sourceforge.net/manual/pythonbindings.html). (I have no idea if that python library is any good, I just googled for Python MFCC). – Richante Apr 10 '12 at 14:22

1 Answers1

0

Have you thought on using an Speech Recognition engine as a basis for your project? You can, for instance, write python code to run CMU sphinx (an BSD option) http://sphinx.subwiki.com/sphinx/index.php/Sphinx3_python_quickstart . Don't know what is you're intention with your fun-project, but if you don't want to study speech recognition algorithms , this could be a good option.

Nemeth
  • 1,070
  • 1
  • 11
  • 16