A model of attention-driven scene analysis

Malcolm Slaney; Trevor Agus; Shih Chii Liu; Merve Kaya; Mounya Elhilali

doi:10.1109/ICASSP.2012.6287838

A model of attention-driven scene analysis

Malcolm Slaney^*, Trevor Agus, Shih Chii Liu, Merve Kaya, Mounya Elhilali

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

5 Citations (Scopus)

Abstract

Parsing complex acoustic scenes involves an intricate interplay between bottom-up, stimulus-driven salient elements in the scene with top-down, goal-directed, mechanisms that shift our attention to particular parts of the scene. Here, we present a framework for exploring the interaction between these two processes in a simulated cocktail party setting. The model shows improved digit recognition in a multi-talker environment with a goal of tracking the source uttering the highest value. This work highlights the relevance of both data-driven and goal-driven processes in tackling real multi-talker, multi-source sound analysis.

Original language	English
Title of host publication	2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings
Pages	145-148
Number of pages	4
DOIs	https://doi.org/10.1109/ICASSP.2012.6287838
Publication status	Published - 23 Oct 2012
Externally published	Yes
Event	IEEE International Conference on Acoustics, Speech, and Signal Processing 2012 - Kyoto, Japan Duration: 25 Mar 2012 → 30 Mar 2012 https://doi.org/10.1109/ICASSP15465.2012

Conference

Conference	IEEE International Conference on Acoustics, Speech, and Signal Processing 2012
Abbreviated title	ICASSP'2012
Country/Territory	Japan
City	Kyoto
Period	25/03/2012 → 30/03/2012
Internet address	https://doi.org/10.1109/ICASSP15465.2012

Keywords

Attention
Auditory Scene Analysis
Cognition
Digit Recognition
Saliency

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP.2012.6287838

Cite this

@inproceedings{e5c3f21052f84209b5db3b5788490842,

title = "A model of attention-driven scene analysis",

abstract = "Parsing complex acoustic scenes involves an intricate interplay between bottom-up, stimulus-driven salient elements in the scene with top-down, goal-directed, mechanisms that shift our attention to particular parts of the scene. Here, we present a framework for exploring the interaction between these two processes in a simulated cocktail party setting. The model shows improved digit recognition in a multi-talker environment with a goal of tracking the source uttering the highest value. This work highlights the relevance of both data-driven and goal-driven processes in tackling real multi-talker, multi-source sound analysis.",

keywords = "Attention, Auditory Scene Analysis, Cognition, Digit Recognition, Saliency",

author = "Malcolm Slaney and Trevor Agus and Liu, {Shih Chii} and Merve Kaya and Mounya Elhilali",

year = "2012",

month = oct,

day = "23",

doi = "10.1109/ICASSP.2012.6287838",

language = "English",

isbn = "9781467300469",

pages = "145--148",

booktitle = "2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings",

note = "IEEE International Conference on Acoustics, Speech, and Signal Processing 2012, ICASSP'2012 ; Conference date: 25-03-2012 Through 30-03-2012",

url = "https://doi.org/10.1109/ICASSP15465.2012",

}

Slaney, M, Agus, T, Liu, SC, Kaya, M & Elhilali, M 2012, A model of attention-driven scene analysis. in 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings., 6287838, pp. 145-148, IEEE International Conference on Acoustics, Speech, and Signal Processing 2012, Kyoto, Japan, 25/03/2012. https://doi.org/10.1109/ICASSP.2012.6287838

TY - GEN

T1 - A model of attention-driven scene analysis

AU - Slaney, Malcolm

AU - Agus, Trevor

AU - Liu, Shih Chii

AU - Kaya, Merve

AU - Elhilali, Mounya

PY - 2012/10/23

Y1 - 2012/10/23

N2 - Parsing complex acoustic scenes involves an intricate interplay between bottom-up, stimulus-driven salient elements in the scene with top-down, goal-directed, mechanisms that shift our attention to particular parts of the scene. Here, we present a framework for exploring the interaction between these two processes in a simulated cocktail party setting. The model shows improved digit recognition in a multi-talker environment with a goal of tracking the source uttering the highest value. This work highlights the relevance of both data-driven and goal-driven processes in tackling real multi-talker, multi-source sound analysis.

AB - Parsing complex acoustic scenes involves an intricate interplay between bottom-up, stimulus-driven salient elements in the scene with top-down, goal-directed, mechanisms that shift our attention to particular parts of the scene. Here, we present a framework for exploring the interaction between these two processes in a simulated cocktail party setting. The model shows improved digit recognition in a multi-talker environment with a goal of tracking the source uttering the highest value. This work highlights the relevance of both data-driven and goal-driven processes in tackling real multi-talker, multi-source sound analysis.

KW - Attention

KW - Auditory Scene Analysis

KW - Cognition

KW - Digit Recognition

KW - Saliency

UR - http://www.scopus.com/inward/record.url?scp=84867606540&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2012.6287838

DO - 10.1109/ICASSP.2012.6287838

M3 - Conference contribution

AN - SCOPUS:84867606540

SN - 9781467300469

SP - 145

EP - 148

BT - 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings

T2 - IEEE International Conference on Acoustics, Speech, and Signal Processing 2012

Y2 - 25 March 2012 through 30 March 2012

ER -

A model of attention-driven scene analysis

Abstract

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this