[Colloquium] TTI-C Talk: Joseph Keshet, IDIAP Research Institute
Julia MacGlashan
macglashan at tti-c.org
Mon Mar 16 15:43:38 CDT 2009
When: Monday, March 23rd @ 11:00am (lunch will be provided after
talk)
Where: 6045 S Kenwood Ave, TTI-C Conference Room #526 (5th Floor)
Who: Joseph Keshet (IDIAP Research Institute, Switzerland)
Title: Discriminative Spoken keyword Detection
The current state-of-the-art automatic speech recognizers are mostly based
on hidden Markov models (HMMs). Despite their popularity, HMM-based
approaches have several known drawbacks such as training objective which is
not aimed at optimizing the evaluation objective. We proposes a new approach
for spoken keyword spotting, which is based on large margin and kernel
methods rather than on HMMs. Unlike previous approaches, the proposed method
employs a discriminative learning procedure, in which the learning phase
aims at achieving a high area under the ROC curve, as this quantity is the
most common measure to evaluate keyword spotters.
The keyword spotter we devise is based on mapping the input acoustic
representation of the speech utterance along with the target keyword into a
vector space. Building on techniques used for large margin and kernel
methods for predicting whole sequences, our keyword spotter distills to a
classifier in this vector-space, which separates speech utterances in which
the keyword is uttered from speech utterances in which the keyword is not
uttered. We describe a simple iterative algorithm for training the keyword
spotter and discuss its formal properties, showing theoretically that it
attains high area under the ROC curve. Experimental results suggest that on
variety standard speech recognition datasets our discriminative system
outperforms the conventional context-independent HMM-based system.
Contact: Karen Livescu, TTI-C klivescu at tti-c.org
834-2549
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20090316/38ea1fce/attachment.htm
More information about the Colloquium
mailing list