Efficient speaker identification and retrieval

  title={Efficient speaker identification and retrieval},
  author={Hagai Aronowitz and David Burshtein},
In this paper we present techniques for efficient s peaker recognition of a large population of speakers and f or e ficient speaker retrieval in large audio archives. We deal with aspects of both time and storage. We use Gaussian mixture m odeling (GMM) for representing both train and test sessions a d show how to perform speaker recognition and retrieval ef fici ntly with only a small degradation in accuracy compared to classic GMM based recognition. We present techniques for ac hieving a… CONTINUE READING


Publications referenced by this paper.
Showing 1-10 of 10 references

Speaker indexing in audio archives using test utterance Gau ssi n mixture modeling

  • H. Aronowitz, D. Burshtein, A. Amir
  • in Proc. of ICSLP,
  • 2005
1 Excerpt

Sp eaker verification using adapted Gaussian mixture models

  • D A.Reynolds, T F.Quatieri, R B.Dunn
  • Digital Signal Processing,
  • 2000
3 Excerpts

A s tudy of computation speed-ups of the GMM-UBM speaker recognition system

  • J. McLaughlin, D A.Reynolds, T. Gleason
  • in Proc. Eurospeech,
  • 1999
3 Excerpts

" Modeling Intra - peaker Variability for Speaker Recognition " , in Proc . of nterspeech , 2005 . eynolds D . A . , " Model compression for GMM based peaker recognition systems "

  • H. ronowitz, D. Burshtein, A Amir, A. Amir, M. Carey, D. Irony

A study f computation speed - ups of the GMM - UBM speaker ecognition system "

  • J. cLaughlin, A. ReynoldsD., T. Gleason

Similar Papers

Loading similar papers…