READING AUDITORY LAB
Anthony J. Watkins
Links:
Research
Some current
research collaborations:
Perceptual constancy
in real-room listening
Some
recent conferences:
Poster and abstract
for Acoustical Society of America, San Antonio, USA 2009
Posters 1 & 2 for British Society of Audiology, Hearing
and Deafness, Southampton 2009
Poster and Chapter (preprint) for
International Symposium on Auditory and Audiological
Research, Helsingør,
Denmark 2009
Chapter (preprint) for
International Symposium on Hearing, Salamanca, Spain 2009
Poster
for EPSRC People in Systems Theme Day, London 2009
Abstract for European
Acoustical Association w/ Acoustical Society of America, Paris 2008
Chapter
for International Symposium on Hearing, Cloppenburg,
Germany 2006
Poster
and abstract for British Society of Audiology,
Hearing and Deafness, Cardiff 2005
Poster
and abstract for British
Society of Audiology, Hearing and Deafness, UCL 2004
First year teaching
Handouts and slides for
4 part 1 lectures on PY1PC: auditory perception
Handouts and slides for
5 part 1 lectures on PY1PL: auditory perception
multiple choice questions - first
year topics in auditory perception
Second year
teaching
Hearing lectures for the part 2 course, PY2PN
Perception: reading list with links to lecture
handouts and slides
Multiple
choice questions – second year topics in auditory perception
Second year course on
Auditory Perception, PY308, reading list and links to lecture handouts
Handouts for PY501: The measurement of sensitivity, Lecture 1 & slides, Lecture 2 & slides, Lecture 3 & slides, TSD
calculations summary, exercise
questions
Handouts and sounds for part
2 miniproject; What affects
the McGurk effect?
Third year teaching
Third year seminars on Auditory Perception, PY3AP: Schedule, reading list, example questions and slides
MSc teaching
MSc seminar; Auditory scene
analysis: List
of demonstrations, reading list, example questions and slides
Sample publications
Watkins, A.J. (1978) "Psychoacoustical
aspects of synthesized vertical locale cues," Journal of the Acoustical
Society of America 63, 1152-1165.
Watkins, A.J. (1985)
"Scale, key and contour in the discrimination of tuned and mistuned
approximations to melody," Perception and Psychophysics 37, 275-285.
Watkins,
A.J. (1988)
"Spectral Transitions and Perceptual compensation for effects of transmission
channels." in Ainsworth, W. and Holmes, J. (eds) Proceedings of the 7th
Symposium of the Federation of Acoustical Societies of Europe: Speech '88. 711-718.
Watkins, A.J. & Makin,
S.J. (1996) "Effects of spectral contrast on perceptual compensation for
spectral-envelope distortion," Journal of the Acoustical Society of
America 99, 3749-3757.
Watkins, A.J. (1999) "The influence of early
reflections on the identification and lateralization of vowels," Journal
of the Acoustical Society of America 106, 2933-2944
Watkins, A.J. & Holt, N.J. (2000) "Effects of
a complex reflection on vowel identification," Acustica
- acta acustica 86,
532-542.
Watkins, A. J. (2008)
"Which is the most common password on the internet?" The Guardian, 29
Jan, G2, p20.
Watkins, A. J., Makin, S.
J. and Raimond, A. P. (2010) Constancy in the perception of speech when the level of
room-reflections varies. In: Buchholz, J., Dau,
T., Dalsgaard, J. and Poulsen,
T. (eds.) Binaural processing and spatial hearing. ISAAR - International
Symposium on Auditory and Audiological Research . The Danavox Jubilee
Foundation, Ballerup, Denmark, pp. 371-380. ISBN
8799001322
Current Research; Speech, Music,
Spatial Hearing and the Auditory Environment
Perception of sound sources seems resistant to diverse
kinds of distortion of the sound by the auditory environment. However, the
perceptual mechanisms responsible are not well understood. A currently puzzling
perceptual mechanism is one demonstrated in my experiments with speech. This is
a mechanism that compensates for 'coloration' effects, which arise from
spectral envelope distortion. As coloration effects are usually present during
listening it is important to understand the compensation mechanism before the
flexibility and robustness of human hearing can be approached by devices that
'listen'. These devices include speaker identifiers, as well as devices that
help others to listen, such as signal processing hearing aids.
A transmission channel is the path from the source to
the listener, and by boosting some frequencies and attenuating others,
coloration arises as the sound passes through the channel. The channel might be
a telephone line, a room that introduces reflections of sound, a hearing aid,
or some combination of these. The pattern of coloration varies considerably
both within and across channels. The channels distort important identifying
characteristics, as many of these characteristics are in the spectral envelopes
of sounds. Such characteristics are crucial for distinctions between spoken
words such as "itch" and "etch". Binaural hearing might
ameliorate the adverse effects of coloration, and current work in this
laboratory is exploring this and other possibilities.
A related project asks why the moderate levels of
reverberation that are present in everyday listening do not seem substantially
to impair the effectiveness of perceptual cues from the amplitude envelope of
speech. These cues help us distinguish between spoken words such as
"sir" and "stir". This project's experiments measure
perception of amplitude-envelope cues while the directional properties of the
reverberant listening conditions are varied. In this way the role of binaural
and other hearing mechanisms can be assessed. The materials used are
realistically complex so that the results relate clearly to the problems
encountered in everyday listening-conditions. The speech stimuli are derived
from recordings of natural utterances, and the reverberation patterns are
obtained from real rooms to create virtual auditory spaces for the listeners.
Here is a speech signal that's been processed with reverberation patterns from
a real room: the first pattern was obtained with a source close to the listener
while the second used a source that was farther away.
nearby source, aiff
or wav
distant source, aiff or wav
unprocessed, aiff or wav
I work for: