14th International Congress of Phonetic Sciences (ICPhS-14)

San Francisco, CA, USA
August 1-7, 1999

A New SNR-Feature Mapping for Robust Multistream Speech Recognition

Frédéric Berthommier (1), Hervé Glotin (1,2)

(1) Institut de la Communication Parlée/INPG, Grenoble, France
(2) IDIAP, Martigny, Switzerland

We describe a new model of CASA labelling which assigns to each time-frequency region a probability "clean" enough to feed a multistream recogniser only adapted to clean data. This labelling process is based on the harmonicity of the speech. The probability is evaluated according to a SNR-feature mapping and the choice of a SNR decision threshold. This allows an extension of a previous method [1] based on the binary detection of noisy time-frequency regions, followed by partial recognition of clean regions. The labelling process i s adapted to a new multistream recognition approach [2], since the previous probabilities serve to weight the streams' posteriors.


  1. Berthommier, F., Glotin, H., Tessier, E., Bourlard, H. (1998) Interfacing of CASA and partial recognition based on a multistream technique, Proc. ICSLP'98, Sydney
  2. Hagen, A., Morris, A. & Bourlard, H. (1998) Subband-Based Speech Recognition in Noisy Conditions: The Full Combination Approach, Res. Report IDIAP, 15, Dec. 98.

Full Paper

Bibliographic reference.  Berthommier, Frédéric / Glotin, Hervé (1999): "A new SNR-feature mapping for robust multistream speech recognition", In ICPhS-14, 711-714.