/AudioMLProject2

Speaker Detection: We have one target speaker whom we want to detect. The set of impostors is open, i.e., we have no prior information on the test speakers. An important part of the speaker detection pipeline is voice activity detection to filter out segments of the signal that do not contain speech.

Primary LanguagePython

Speaker Detection – from GMMs to i-vectors

Goals

  • Mix corpora
  • Create a GMM based speaker detection system with MAP adaption
  • Create an i-vector based speaker detection system
  • Evaluate
  • Present your results in form of a report