Evaluating the Audio-Visual Speech Enhancement Challenge (AVSEC) Baseline Model Using an Out-of-Domain Free-Flowing Corpus

Kia K. Dashtipour; Mandar Gogate; Adeel Hussain; Bryony Buck; Arif Reza Anwary; Tughrul Arslan; Amir Hussain

doi:10.21437/AVSEC.2024-16

Evaluating the Audio-Visual Speech Enhancement Challenge (AVSEC) Baseline Model Using an Out-of-Domain Free-Flowing Corpus

Kia K. Dashtipour (Lead / Corresponding author), Mandar Gogate, Adeel Hussain, Bryony Buck, Arif Reza Anwary, Tughrul Arslan, Amir Hussain

Psychology

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

10 Downloads (Pure)

Abstract

The human auditory cortex contextually integrates audio-visual (AV) cues to enhance the comprehension of speech in noisy environments. Numerous studies have investigated the effectiveness of AV integration for speech enhancement (SE). This paper evaluates the effectiveness of the COG-MHEAR AV SE Challenge baseline model using an out-of-domain free-flowing corpus. Experimental results indicate that the COG-MHEAR AV SE Challenge baseline model exhibits superior performance when applied to an out-of-domain corpus.

Original language	English
Pages	75-78
Number of pages	4
DOIs	https://doi.org/10.21437/AVSEC.2024-16
Publication status	Published - 1 Sept 2024
Event	3rd COG-MHEAR Workshop on Audio-Visual Speech Enhancement - Kos, Greece Duration: 1 Sept 2024 → 1 Sept 2024 https://www.isca-archive.org/avsec_2024/index.html

Workshop

Workshop	3rd COG-MHEAR Workshop on Audio-Visual Speech Enhancement
Country/Territory	Greece
City	Kos
Period	1/09/24 → 1/09/24
Internet address	https://www.isca-archive.org/avsec_2024/index.html

Access to Document

10.21437/AVSEC.2024-16Licence: Other

Final published versionFinal published version, 485 KBLicence: Other

https://www.isca-archive.org/avsec_2024/dashtipour24b_avsec.html

Cite this

@conference{f067c2030de74b438c08895a3f953269,

title = "Evaluating the Audio-Visual Speech Enhancement Challenge (AVSEC) Baseline Model Using an Out-of-Domain Free-Flowing Corpus",

abstract = "The human auditory cortex contextually integrates audio-visual (AV) cues to enhance the comprehension of speech in noisy environments. Numerous studies have investigated the effectiveness of AV integration for speech enhancement (SE). This paper evaluates the effectiveness of the COG-MHEAR AV SE Challenge baseline model using an out-of-domain free-flowing corpus. Experimental results indicate that the COG-MHEAR AV SE Challenge baseline model exhibits superior performance when applied to an out-of-domain corpus.",

author = "Dashtipour, \{Kia K.\} and Mandar Gogate and Adeel Hussain and Bryony Buck and Anwary, \{Arif Reza\} and Tughrul Arslan and Amir Hussain",

year = "2024",

month = sep,

day = "1",

doi = "10.21437/AVSEC.2024-16",

language = "English",

pages = "75--78",

note = "3rd COG-MHEAR Workshop on Audio-Visual Speech Enhancement ; Conference date: 01-09-2024 Through 01-09-2024",

url = "https://www.isca-archive.org/avsec\_2024/index.html",

}

Dashtipour, KK, Gogate, M, Hussain, A, Buck, B, Anwary, AR, Arslan, T & Hussain, A 2024, 'Evaluating the Audio-Visual Speech Enhancement Challenge (AVSEC) Baseline Model Using an Out-of-Domain Free-Flowing Corpus', Paper presented at 3rd COG-MHEAR Workshop on Audio-Visual Speech Enhancement, Kos, Greece, 1/09/24 - 1/09/24 pp. 75-78. https://doi.org/10.21437/AVSEC.2024-16

Evaluating the Audio-Visual Speech Enhancement Challenge (AVSEC) Baseline Model Using an Out-of-Domain Free-Flowing Corpus. / Dashtipour, Kia K. (Lead / Corresponding author); Gogate, Mandar; Hussain, Adeel et al.
2024. 75-78 Paper presented at 3rd COG-MHEAR Workshop on Audio-Visual Speech Enhancement, Kos, Greece.

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

TY - CONF

T1 - Evaluating the Audio-Visual Speech Enhancement Challenge (AVSEC) Baseline Model Using an Out-of-Domain Free-Flowing Corpus

AU - Dashtipour, Kia K.

AU - Gogate, Mandar

AU - Hussain, Adeel

AU - Buck, Bryony

AU - Anwary, Arif Reza

AU - Arslan, Tughrul

AU - Hussain, Amir

PY - 2024/9/1

Y1 - 2024/9/1

N2 - The human auditory cortex contextually integrates audio-visual (AV) cues to enhance the comprehension of speech in noisy environments. Numerous studies have investigated the effectiveness of AV integration for speech enhancement (SE). This paper evaluates the effectiveness of the COG-MHEAR AV SE Challenge baseline model using an out-of-domain free-flowing corpus. Experimental results indicate that the COG-MHEAR AV SE Challenge baseline model exhibits superior performance when applied to an out-of-domain corpus.

AB - The human auditory cortex contextually integrates audio-visual (AV) cues to enhance the comprehension of speech in noisy environments. Numerous studies have investigated the effectiveness of AV integration for speech enhancement (SE). This paper evaluates the effectiveness of the COG-MHEAR AV SE Challenge baseline model using an out-of-domain free-flowing corpus. Experimental results indicate that the COG-MHEAR AV SE Challenge baseline model exhibits superior performance when applied to an out-of-domain corpus.

U2 - 10.21437/AVSEC.2024-16

DO - 10.21437/AVSEC.2024-16

M3 - Paper

SP - 75

EP - 78

T2 - 3rd COG-MHEAR Workshop on Audio-Visual Speech Enhancement

Y2 - 1 September 2024 through 1 September 2024

ER -

Evaluating the Audio-Visual Speech Enhancement Challenge (AVSEC) Baseline Model Using an Out-of-Domain Free-Flowing Corpus

Abstract

Workshop

Access to Document

Fingerprint

Cite this