I'm trying to create my own program for speaker identification (text-independent) and have used MFCC features for that. But result of recognition is very poor. My supervisor told me that using MFCC for such problem is not the best solution because we are loosing a lot of data about speaker and MFCC mostly used for speech recognition (what is said, not who).

Does anyone know which features can give better result for speaker recognition?


  • $\begingroup$ Use a deep neural network to learn the features. $\endgroup$ – Emre Nov 27 '14 at 21:56
  • $\begingroup$ MFCCs are used for both speech and speaker recognition. Yes! from the first look, it may look very weird, but still MFCCs are used in conventional as well as state-of-the art system. You may add some post-processing techniques such as rasta, CMVN to enhance the features further. $\endgroup$ – talk2speech Jan 27 '16 at 22:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.