Speaker-independent machine lip-reading with speaker-dependent viseme classifiers

Helen L. Bear; Stephen Cox; Richard Harvey

Speaker-independent machine lip-reading with speaker-dependent viseme classifiers

Helen L. Bear, Stephen Cox, Richard Harvey

Research output: Contribution to conference › Paper › peer-review

Abstract

In machine lip-reading, which is identification of speech from visual-only information, there is evidence to show that visual speech is highly dependent upon the speaker (Cox et al, 2008). Here, we use a phoneme-clustering method to form new phoneme-to-viseme maps for both individual and multiple speakers. We use these maps to examine how similarly speakers talk visually. We conclude that broadly speaking, speakers have the same repertoire of mouth gestures, where they differ is in the use of the gestures.

Original language	English
Publication status	Published - Sep 2015
Event	FAAVSP - The 1st Joint Conference on Facial Analysis, Animation and Auditory-Visual Speech Processing - Austria, Vienna, Austria Duration: 11 Sep 2015 → 13 Sep 2015 http://www.isca-speech.org/archive/avsp15/av15_127.html

Conference

Conference	FAAVSP - The 1st Joint Conference on Facial Analysis, Animation and Auditory-Visual Speech Processing
Abbreviated title	FAAVSP 2015
Country/Territory	Austria
City	Vienna
Period	11/09/15 → 13/09/15
Internet address	http://www.isca-speech.org/archive/avsp15/av15_127.html

Richard Harvey
- R.W.Harveyuea.acuk
- School of Computing Sciences - Professor
Person: Research Group Member, Academic, Teaching & Research

Cite this

@conference{fba219e16bfa41669275ea980fa376da,

title = "Speaker-independent machine lip-reading with speaker-dependent viseme classifiers",

abstract = "In machine lip-reading, which is identification of speech from visual-only information, there is evidence to show that visual speech is highly dependent upon the speaker (Cox et al, 2008). Here, we use a phoneme-clustering method to form new phoneme-to-viseme maps for both individual and multiple speakers. We use these maps to examine how similarly speakers talk visually. We conclude that broadly speaking, speakers have the same repertoire of mouth gestures, where they differ is in the use of the gestures.",

author = "Bear, {Helen L.} and Stephen Cox and Richard Harvey",

year = "2015",

month = sep,

language = "English",

note = "FAAVSP - The 1st Joint Conference on Facial Analysis, Animation and Auditory-Visual Speech Processing, FAAVSP 2015 ; Conference date: 11-09-2015 Through 13-09-2015",

url = "http://www.isca-speech.org/archive/avsp15/av15_127.html",

}

TY - CONF

T1 - Speaker-independent machine lip-reading with speaker-dependent viseme classifiers

AU - Bear, Helen L.

AU - Cox, Stephen

AU - Harvey, Richard

PY - 2015/9

Y1 - 2015/9

N2 - In machine lip-reading, which is identification of speech from visual-only information, there is evidence to show that visual speech is highly dependent upon the speaker (Cox et al, 2008). Here, we use a phoneme-clustering method to form new phoneme-to-viseme maps for both individual and multiple speakers. We use these maps to examine how similarly speakers talk visually. We conclude that broadly speaking, speakers have the same repertoire of mouth gestures, where they differ is in the use of the gestures.

AB - In machine lip-reading, which is identification of speech from visual-only information, there is evidence to show that visual speech is highly dependent upon the speaker (Cox et al, 2008). Here, we use a phoneme-clustering method to form new phoneme-to-viseme maps for both individual and multiple speakers. We use these maps to examine how similarly speakers talk visually. We conclude that broadly speaking, speakers have the same repertoire of mouth gestures, where they differ is in the use of the gestures.

M3 - Paper

T2 - FAAVSP - The 1st Joint Conference on Facial Analysis, Animation and Auditory-Visual Speech Processing

Y2 - 11 September 2015 through 13 September 2015

ER -

Speaker-independent machine lip-reading with speaker-dependent viseme classifiers

Abstract

Conference

Profiles

Richard Harvey

Cite this