TCS Project -Distant speaker recognition using microphone array
Electrical Engineering Dept. IIT Bombay, Powai
Past work
Source localization and beamforming was done on the available TCS data (non-overlapping,partially overlappping anf fully overlapping speakers). Objective meansures were obtained for various beamforming methods for non-overlapping speakers.
Presentations
Some of our presentations/slides during the discussions.
Relevant websites
- Beamformit can be obtained from github
- REVERB challenge
- A multi-microphone signal processing for speech enhancement completed at UC Berkeley (ICSI). Has information on meetings recorder digits (MRD) data. Also relates to the Beamformit tool.
- RT 09 challenge NIST evaluation
- SiSEC challenge
References
- C. Zhang, D. Florêncio, D. E. Ba, and Z. Zhang, “Maximum likelihood sound source localization and beamforming for directional microphone arrays in distributed meetings,” IEEE Transactions on Multimedia, vol. 10, no. 3, pp. 538–548, 2008.[download]
- C. Knapp and G. Carter, “The generalized correlation method for estimation of time delay,” IEEE Transactions on Acoustics, Speech, and Signal Processing, vol. 24, no. 4, pp. 320–327, 1976.[download]
- Xavier Anguera, Chuck Wooters and Javier Hernando, Acoustic beamforming for speaker diarization of meetings, IEEE Trans. Audio, Speech, and Lang. Proc., Sep. 2007, vol. 15, no. 7, pp. 2011-2023.[download]
- Xavier Anguera, Robust Speaker Diarization for Meetings, PhD Thesis, UPC Barcelona, 2006. [download]
- Xavier Anguera, Miro Simon Bozonnet, Nicholas Evans, Corinne Fredouille, Gerald Friedland, and Oriol Vinyals, Speaker Diarization: A Review of Recent Research, IEEE Trans. Audio, Speech, and Lang. Proc., Feb. 2012, vol. 20, no. 2, pp. 356-370.[download]