logo Idiap Research Institute        
 [BibTeX] [Marc21]
Exploiting Contextual Information for Speech/Non-Speech Detection
Type of publication: Conference paper
Citation: Parthasarathi_TSD2008_2008
Booktitle: Text, Speech and Dialogue
Series: Series of Lecture Notes In Artificial Intelligence (LNAI)
Volume: 5246
Year: 2008
Month: 9
Publisher: Springer-Verlag Berlin, Heidelberg
Location: Brno, Czech Republic
ISBN: 978-3-540-87390-7
Crossref: parthasarathi:rr08-22:
Abstract: In this paper, we investigate the effect of temporal context for speech/non-speech detection (SND). It is shown that even a simple feature such as full-band energy, when employed with a large-enough context, shows promise for further investigation. Experimental evaluations on the test data set, with a state-of-the-art multi-layer perceptron based SND system and a simple energy threshold based SND method, using the F-measure, show an absolute performance gain of 4.4% and 5.4% respectively. The optimal contextual length was found to be 1000 ms. Further numerical optimizations yield an improvement (3.37% absolute,',','), resulting in an absolute gain of 7.77% and 8.77% over the MLP based and energy based methods respectively. ROC based performance evaluation also reveals promising performance for the proposed method, particularly in low SNR conditions.
Keywords:
Projects Idiap
Authors Parthasarathi, Sree Hari Krishnan
Motlicek, Petr
Hermansky, Hynek
Added by: [UNK]
Total mark: 0
Attachments
  • Parthasarathi_TSD2008_2008.pdf
Notes