Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385001640> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4385001640 abstract "Automatic speech recognition (ASR) systems are designed to transcribe spoken language into written text and find utility in a variety of applications including voice assistants and transcription services. However, it has been observed that state-of-the-art ASR systems which deliver impressive benchmark results, struggle with speakers of certain regions or demographics due to variation in their speech properties. In this work, we describe the curation of a massive speech dataset of 8740 hours consisting of $sim9.8$K technical lectures in the English language along with their transcripts delivered by instructors representing various parts of Indian demography. The dataset is sourced from the very popular NPTEL MOOC platform. We use the curated dataset to measure the existing disparity in YouTube Automatic Captions and OpenAI Whisper model performance across the diverse demographic traits of speakers in India. While there exists disparity due to gender, native region, age and speech rate of speakers, disparity based on caste is non-existent. We also observe statistically significant disparity across the disciplines of the lectures. These results indicate the need of more inclusive and robust ASR systems and more representational datasets for disparity evaluation in them." @default.
- W4385001640 created "2023-07-22" @default.
- W4385001640 creator A5010500831 @default.
- W4385001640 creator A5020991141 @default.
- W4385001640 creator A5054177518 @default.
- W4385001640 date "2023-07-20" @default.
- W4385001640 modified "2023-10-17" @default.
- W4385001640 title "A Deep Dive into the Disparity of Word Error Rates Across Thousands of NPTEL MOOC Videos" @default.
- W4385001640 doi "https://doi.org/10.48550/arxiv.2307.10587" @default.
- W4385001640 hasPublicationYear "2023" @default.
- W4385001640 type Work @default.
- W4385001640 citedByCount "0" @default.
- W4385001640 crossrefType "posted-content" @default.
- W4385001640 hasAuthorship W4385001640A5010500831 @default.
- W4385001640 hasAuthorship W4385001640A5020991141 @default.
- W4385001640 hasAuthorship W4385001640A5054177518 @default.
- W4385001640 hasBestOaLocation W43850016401 @default.
- W4385001640 hasConcept C121332964 @default.
- W4385001640 hasConcept C13280743 @default.
- W4385001640 hasConcept C136197465 @default.
- W4385001640 hasConcept C137293760 @default.
- W4385001640 hasConcept C138885662 @default.
- W4385001640 hasConcept C144024400 @default.
- W4385001640 hasConcept C149923435 @default.
- W4385001640 hasConcept C154945302 @default.
- W4385001640 hasConcept C185798385 @default.
- W4385001640 hasConcept C204321447 @default.
- W4385001640 hasConcept C205649164 @default.
- W4385001640 hasConcept C2778334786 @default.
- W4385001640 hasConcept C2779509574 @default.
- W4385001640 hasConcept C2780084366 @default.
- W4385001640 hasConcept C28490314 @default.
- W4385001640 hasConcept C40969351 @default.
- W4385001640 hasConcept C41008148 @default.
- W4385001640 hasConcept C41895202 @default.
- W4385001640 hasConcept C44870925 @default.
- W4385001640 hasConcept C90805587 @default.
- W4385001640 hasConceptScore W4385001640C121332964 @default.
- W4385001640 hasConceptScore W4385001640C13280743 @default.
- W4385001640 hasConceptScore W4385001640C136197465 @default.
- W4385001640 hasConceptScore W4385001640C137293760 @default.
- W4385001640 hasConceptScore W4385001640C138885662 @default.
- W4385001640 hasConceptScore W4385001640C144024400 @default.
- W4385001640 hasConceptScore W4385001640C149923435 @default.
- W4385001640 hasConceptScore W4385001640C154945302 @default.
- W4385001640 hasConceptScore W4385001640C185798385 @default.
- W4385001640 hasConceptScore W4385001640C204321447 @default.
- W4385001640 hasConceptScore W4385001640C205649164 @default.
- W4385001640 hasConceptScore W4385001640C2778334786 @default.
- W4385001640 hasConceptScore W4385001640C2779509574 @default.
- W4385001640 hasConceptScore W4385001640C2780084366 @default.
- W4385001640 hasConceptScore W4385001640C28490314 @default.
- W4385001640 hasConceptScore W4385001640C40969351 @default.
- W4385001640 hasConceptScore W4385001640C41008148 @default.
- W4385001640 hasConceptScore W4385001640C41895202 @default.
- W4385001640 hasConceptScore W4385001640C44870925 @default.
- W4385001640 hasConceptScore W4385001640C90805587 @default.
- W4385001640 hasLocation W43850016401 @default.
- W4385001640 hasOpenAccess W4385001640 @default.
- W4385001640 hasPrimaryLocation W43850016401 @default.
- W4385001640 hasRelatedWork W1578024259 @default.
- W4385001640 hasRelatedWork W2019914509 @default.
- W4385001640 hasRelatedWork W2786018489 @default.
- W4385001640 hasRelatedWork W2884483159 @default.
- W4385001640 hasRelatedWork W2953291251 @default.
- W4385001640 hasRelatedWork W3095760691 @default.
- W4385001640 hasRelatedWork W4287629333 @default.
- W4385001640 hasRelatedWork W4290682478 @default.
- W4385001640 hasRelatedWork W4306309629 @default.
- W4385001640 hasRelatedWork W4319779560 @default.
- W4385001640 isParatext "false" @default.
- W4385001640 isRetracted "false" @default.
- W4385001640 workType "article" @default.