ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Acoustic and linguistic features influence talker change detection

Sharma, NK and Krishnamohan, V and Ganapathy, S and Gangopadhayay, A and Fink, L (2020) Acoustic and linguistic features influence talker change detection. In: Journal of the Acoustical Society of America, 148 (5). EL414-EL419.

jou_aco_soc_148-5_EL414-EL419_2020.pdf - Published Version

Download (19MB) | Preview
Official URL: https://doi.org/10.1121/10.0002462


A listening test is proposed in which human participants detect talker changes in two natural, multi-talker speech stimuli sets - a familiar language (English) and an unfamiliar language (Chinese). Miss rate, false-alarm rate, and response times (RT) showed a significant dependence on language familiarity. Linear regression modeling of RTs using diverse acoustic features derived from the stimuli showed recruitment of a pool of acoustic features for the talker change detection task. Further, benchmarking the same task against the state-of-the-art machine diarization system showed that the machine system achieves human parity for the familiar language but not for the unfamiliar language. © 2020 Acoustical Society of America.

Item Type: Journal Article
Publication: Journal of the Acoustical Society of America
Publisher: Acoustical Society of America
Additional Information: The copyright for this article belongs to The Authors.
Keywords: Linguistics, Acoustic features; Change detection; False alarm rate; Linear regression models; Linguistic features; Listening tests; Machine systems; State of the art, Feature extraction, adult; article; benchmarking; female; human; human experiment; language; linear regression analysis; male; parity; reaction time; speech
Department/Centre: Division of Electrical Sciences > Electrical Engineering
Date Deposited: 10 Jan 2023 11:59
Last Modified: 10 Jan 2023 11:59
URI: https://eprints.iisc.ac.in/id/eprint/79027

Actions (login required)

View Item View Item