SANE 2012 - Speech and Audio in the Northeast

October 24, 2012

Host: Jonathan Le Rouxleroux[at]merl[dot]com

SANE 2012

SANE 2012, a one-day event gathering researchers and students in speech and audio from the northeast of the American continent, will be held on Wednesday October 24, 2012 at Mitsubishi Electric Research Laboratories (MERL) in Cambridge, MA.


  • Date: October 24, 2012; 8:30am - 5:00pm
  • Venue: MERL - 201 Broadway, Cambridge, MA 02139 (Access)


8:30-9:00Registration and Breakfast
9:10-9:55Jim Glass and Chia-ying Lee (MIT CSAIL)
"Zero-Resource Speech Pattern and Sub-Word Unit Discovery"
9:55-10:40Tara Sainath (IBM Research)
"Deep Belief Network Research at IBM"
10:40-11:00Coffee Break 1
11:00-11:45Dan Ellis (Columbia University)
"Recognizing and Classifying Environmental Sounds"
11:45-12:30Josh McDermott (MIT BCS)
"Understanding Audition via Sound Analysis and Synthesis"
1:30-2:15Herb Gish (BBN - Raytheon)
"Self-Organizing Units (SOUs): Training Speech Recognizers Without Any Transcribed Audio"
2:15-3:00Timothy J. Hazen (MIT Lincoln Labs) and David Harwath (MIT CSAIL)
"Latent Topic Modeling of Conversational Speech"
3:00-3:20Coffee Break 2
3:20-4:05Steven J. Rennie (IBM Research)
"Factorial Hidden Restricted Boltzmann Machines for Noise Robust Speech Recognition"
4:05-4:50John R. Hershey (MERL)
"A New Class of Dynamical System Models for Speech and Audio"
4:50-5:00Closing remarks


Registration is free but required. We will only be able to accomodate a limited number of participants, so we encourage those interested in attending this event to register as soon as possible by sending an email to sane2012[at]merl[dot]com with your name and affiliation.
Update: registration is now closed, and the waiting list is full.


The workshop will be hosted at MERL, on the 8th floor of 201 Broadway, Cambridge, MA 02139. MERL is located within a short walk from Kendall/MIT station on the T Red Line. Please come directly to the 8th floor lobby to register.

Organizing Committee