Phone Adaptive Training for Speaker Diarization

@inproceedings{Bozonnet2012PhoneAT,
  title={Phone Adaptive Training for Speaker Diarization},
  author={Simon Bozonnet and Ravichander Vipperla and Nicholas W. D. Evans},
  booktitle={INTERSPEECH},
  year={2012}
}
The linguistic content of a speech signal is a source of unwanted variation which can degrade speaker diarization performance. This paper presents our latest work to reduce its impact. The new approach, referred to as Phone Adaptive Training (PAT), is analogous to speaker adaptive training used in automatic speech recognition. We report an oracle experiment which shows that PAT has the potential to deliver a 33% relative improvement in the diarization error rate over our baseline system… CONTINUE READING

From This Paper

Figures, tables, results, and topics from this paper.

Key Quantitative Results

  • We report an oracle experiment which shows that PAT has the potential to deliver a 33% relative improvement in the diarization error rate over our baseline system.

References

Publications referenced by this paper.
SHOWING 1-9 OF 9 REFERENCES

The IIR- NTU Speaker Diarization Systems for RT 2009

  • T. H. Nguyen, H. Sun, +6 authors H. Li
  • RT’09, NIST Rich Transcription Workshop, May 28…
  • 2009
2 Excerpts

A compact model for speaker - adaptive train

  • J. Mcdonough T. Anastasakos, R. Schwartz, J. Makhoul
  • Proc . ICSLP
  • 1996

Similar Papers

Loading similar papers…