BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Time-domain multi-channel speech separation and extraction - Jisi 
 Zhang\, University of Sheffield
DTSTART:20210608T110000Z
DTEND:20210608T120000Z
UID:TALK160780@talks.cam.ac.uk
CONTACT:Dr Kate Knill
DESCRIPTION:*Abstract*: When multiple speakers talk at the same time\, eac
 h utterance is partially or completely overlapped by one or more competing
  utterances. The overlapping speech is challenging to speech technologies\
 , including automatic speech recognition\, speaker diarization\, and speak
 er verification. These challenges can be overcome by using speech separati
 on front-ends\, which aim to segregate individual source speakers from a m
 ixture signal. Despite the recent progress of single-channel speech separa
 tion driven by advances in deep learning\, it still performs poorly in dis
 tant microphone scenarios where noise and reverberation are involved. This
  talk focuses on the development of multi-channel speech separation techni
 ques for separating mixture signals in the distant microphone case. The ta
 lk will be split into three parts. The first part introduces an end-to-end
  neural architecture with time-domain multi-microphone input. Second\, the
  knowledge of speaker identity is exploited to extend the multi-channel se
 paration system to perform a multi-speaker extraction task. Finally\, an u
 nsupervised approach is described\, which aims to applying the end-to-end 
 separation system in situations where supervised data is hard to collect. 
 The methods are evaluated using simulated data with reverberation and ambi
 ent noise\, and in terms of signal enhancement metrics and as front-ends t
 o ASR.\n\n*Bio*: Jisi Zhang is a final year PhD student supervised by Prof
 essor Jon Barker\, in the Department of Computer Science at the University
  of Sheffield. He is interested in speech separation\, multi-channel proce
 ssing\, and multi-talker speech recognition.\n
LOCATION:Zoom: https://zoom.us/j/95352633552?pwd=RzJVK2UzOGZyNU5mVHd1Y1VPT
 2tDUT09
END:VEVENT
END:VCALENDAR