Work

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

Year: 2024

Type: conference-paper

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.

DOI: 10.21437/avsec.2024-16

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

Language: English

Topic: Speech and Audio Processing

Subfield: Signal Processing

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: Royal Society, Royal Society of Edinburgh, Engineering and Physical Sciences Research Council

Awards: Natural Language Generation for Low-resource Domains, COG-MHEAR: Towards cognitively-inspired 5G-IoT enabled, multi-modal Hearing Aids, G386565863

OpenAlex ID: W4402535303

FWCI: 1.143

Cites:

Cited by: 3

Related to: 10