Logistic Discriminative Speech Detection using Posterior SNRs
- Arun C. Surendran ,
- Somsak Sukuttanon ,
- John Platt
Published by Institute of Electrical and Electronics Engineers, Inc.
We introduce an elegant and novel design for a speech detector which estimates the probability of the presence of speech in each time-frequency bin, as well as in each frame. The proposed system uses discriminative estimators based on logistic regression, and incorporates spectral and temporal correlations in the same framework. The detector is flexible enough to be configured in a single level or a “stacked” bi-level architecture depending on the needs of the application. An important part of the proposed design is the use of a new set of features: the normalized logarithm of the estimated posterior signal-to-noise ratio. These can be easily and automatically generated by tracking the noise spectrum online. We present results on the AURORA database to demonstrate that the overall design is simple, flexible and effective.
© 2004 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.