SubtitleFormatter: Making Subtitles Easier to Read for Deaf and Hard of Hearing Viewers on Personal Devices

Kushalnagar, Raja; Kushalnagar, Kesavan

doi:10.1007/978-3-319-94277-3_35

Cited by 3 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DHH viewers are often dissatisfied with the quality of captioning provided in live contexts, which provide less time for caption production than prerecorded contexts (Amin et al, 2021b;Kushalnagar and Kushalnagar, 2018). If regulatory organizations that measure the quality of captions used quality metrics that better reflect the DHH users' preferences, DHH viewers' experience may improve.…”

Section: Introductionmentioning

confidence: 99%

Using BERT Embeddings to Model Word Importance in Conversational Transcripts for Deaf and Hard of Hearing Users

Amin¹,

Hassan²,

Alm³

et al. 2022

Proceedings of the Second Workshop on Language Technology for Equality, Diversity and Inclusion

View full text Add to dashboard Cite

Deaf and hard of hearing individuals regularly rely on captioning while watching live TV. Live TV captioning is evaluated by regulatory agencies using various caption evaluation metrics. However, caption evaluation metrics are often not informed by preferences of DHH users or how meaningful the captions are. There is a need to construct caption evaluation metrics that take the relative importance of words in a transcript into account. We conducted correlation analysis between two types of word embeddings and human-annotated labeled wordimportance scores in existing corpus. We found that normalized contextualized word embeddings generated using BERT correlated better with manually annotated importance scores than word2vec-based word embeddings. We make available a pairing of word embeddings and their human-annotated importance scores. We also provide proof-of-concept utility by training word importance models, achieving an F1-score of 0.57 in the 6-class word importance classification task.

show abstract

Section: Introductionmentioning

confidence: 99%