BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Towards Weaker Supervision and Simpler Pipelines in Speech Recogni
 tion  - Gabriel Synnaeve
DTSTART:20160527T100000Z
DTEND:20160527T110000Z
UID:TALK66366@talks.cam.ac.uk
CONTACT:Louise Segar
DESCRIPTION:raditionally\, speech recognition required feature engineering
 \, fine grained transcription\, a phonetics step\, and glue all of these w
 ith long pipelines. The progress in computational power and datasets size 
 enabled the success of a less rigid class of deep learning models. We will
  present our recent works towards coarser annotation: either training acou
 stic models directly on pairs of same or different words (no class informa
 tion nor phonetic information) with siamese neural networks\, or training 
 on sentences annotated as bag of words with large convolutional neural net
 works and temporal pooling. We will also show promising results on unsuper
 vised acoustic model training. Finally\, we will present results towards t
 raining directly from the raw waveform to the graphemes with an efficient 
 sequence-based loss. This will span joint work with Neil Zeghidour\, Ronan
  Collobert\, Dimitri Palaz\, Nicolas Usunier\, Christian Puhrsch\, and Emm
 anuel Dupoux.
LOCATION:Engineering Department\, CBL Room BE-438
END:VEVENT
END:VCALENDAR