Home

Call for
papers

Dates

Submissions

Committee

Program

Consistent & Reliable Acoustic Cues
for sound analysis

Program

Sunday September 2nd 2001

Session 1 09:00-10:30

PERCEPTION & AUDITORY MODELS

Overview	Phil Green
Oral	Johannes Nix & Volker Hohmann Enhancing Sound Sources by use of Binaural Spatial Cues
	Alain de Cheveigné Generalized Correlation Network model of auditory processing
Posters	Toshio Irino, R. D. Patterson, and H. Kawahara Sound resynthesis from Auditory Mellin Image using STRAIGHT
	Olivier Crouzet & W.A. Ainsworth On the various influences of envelope information on the perception of speech in adverse conditions: An analysis of between-channel envelope correlation
	Philip J.B. Jackson Acoustic cues of voiced and voiceless plosives for determining place of articulation
	Matti Karjalainen Auditory Interpretation and Application of Warped Linear Prediction
	Shuangyu Chang, Lokendra Shastri & Steven Greenberg Robust Phonetic Feature Extraction Under a Wide Range of Noise Backgrounds and Signal-to-Noise Ratios

Session 2 11:00-12:30

MUSIC & GENERAL AUDIO ANALYSIS

Overview	Dan Ellis
Oral	Anssi Klapuri, Tuomas Virtanen, Antti Eronen & Jarno Seppänen Automatic transcription of musical recordings
	Michael Casey Reduced-Rank Spectra and Entropic Priors as Consistent and Reliable Cues for Generalized Sound Recognition
Posters	Silvia Allegro & Stefan Launer Sound Classification in Hearing Instruments by means of Auditory Scene Analysis
	Shoko Araki, Shoji Makino, Ryo Mukai & Hiroshi Saruwatari Equivalence between Frequency Domain Blind Source Separation and Frequency Domain Adaptive Beamformers
	Tomoya Narita & Masahide Sugiyama Fast Music Retrieval using Spectrum and Power Information
	Shin'ichi Takeuchi, Masaki Yamashita, Takayuki Uchida, Masahide Sugiyama Optimization of Voice/Music Detection in Sound Data
	Masataka Goto A Predominant-F0 Estimation Method for Real-world Musical Audio Signals: MAP Estimation for Incorporating Prior Knowledge about F0s and Tone Models
	Daniel P.W. Ellis Detecting alarm sounds

Session 3 14:00-15:30

MISSING-DATA SPEECH RECOGNITION

Overview	Martin Cooke
Oral	Jon Barker, Martin Cooke & Dan Ellis Integrating bottom-up and top-down constraints to achieve robust ASR: The multisource decoder
	Phillippe Renevey & Andrzej Drygajlo Detection of Reliable Features for Speech Recognition in Noisy Conditions Using a Statistical Criterion
Posters	Kalle Palomäki, Guy Brown & DeLiang Wang A Binaural Model for Missing Data Speech Recognition in Noisy and Reverberant Conditions
	Juha Häkkinen & Hemmo Haverinen On the Use of Missing Feature Theory with Cepstral Features
	A. C. Morris Data Utility Modelling for Mismatch Reduction
	Herve Glotin Robust multi-stream speech recognition based on the combined reliabilities of the speech signal (voicing cue) and phonemes estimates using a bias prediction
	Stéphane Dupont & Christophe Ris Multiband with contaminated training data
	Bhiksha Raj, Michael L. Seltzer & Richard M. Stern Robust Speech Recognition using Missing Features: the Case for Restoring Missing Input Features

Session 4 16:00-17:30

APPROACHES TO HANDLING NOISY SPEECH

Overview	Hiroshi G Okuno
Oral	Emmanuel Tessier & Frédéric Berthommier Speech enhancement and segregation based on the localisation cue for cocktail-party processing
	Ikuyo Masuda-Katsuse & Yoshimori Sugano Speech estimation biased by phonemic expectation in the presence of non-stationary and unpredictable noise
Posters	Laurens van de Werff, Johan de Veth, Bert Cranen & Louis Boves Analysis of Disturbed Acoustic Features in terms of Emission Cost
	Yuichi ISHIMOTO, Masashi UNOKI & Masato AKAGI A fundamental frequency estimation method for noisy speech based on instantaneous amplitude and frequency
	Joan Marí , José Manuel Ferrer & Fritz Class Evaluation of Robust Feature Extraction and Acoustic Modelling algorithms/systems by interfacing ASR systems
	Hiroshi G. Okuno, Kazuhiro Nakadai & Hiroaki Kitano Effects of increasing modalities in understanding three simultaneous speeches with two microphones
	Yasuhiro Minami, Erik McDermott, Atsushi Nakamura & Shigeru Katagiri A recognition method using synthesis-based scoring that incorporates direct relations between static and dynamic feature vector time series

This is a paragraph of dummy text whose purpose is to see if that by adding a line of text long enough to span more than the entire window we can in fact persuade the table rendering algorithms to respect the width=pixels specification of the leftmost column as an absolute maximum, rather than stretching it when the rightmost column is thinner.