Leveraging Multiple Sources in Automatic African American English Dialect Detection for Adults and Children
Domain:
natural language processingRecord type:
paper
Creator:
Ale
Publisher:
Host:
IEEE ICASSP 2023 Conference, Hybrid Event, 4-10 June 2023, Rhodes Island, Greece This work presents a novel system which utilizes acoustic, phonological, morphosyntactic, and prosodic information for binary automatic dialect detection of African American English. We train this system utilizing out-of-domain adult speech data and then evaluate on both children's and adults' speech with unmatched training and testing scenarios. The proposed system combines novel and state-of-the-art architectures, including a multi-modal masked language model pre-trained on Twitter text data and fine-tuned on ASR transcripts as well as an LSTM trained on self-supervised learning representations, in order to learn a comprehensive view of dialect. We show robust performance for our system across age and recording conditions.
Similar
Automatic Dialect Density Estimation for African American EnglishDialect-Specific Models for Automatic Speech Recognition of African American Vernacular EnglishDialect Awareness and Lexical Comprehension of Mainstream American English in African American English–Speaking ChildrenPhonological Milestones for African American English-Speaking Children Learning Mainstream American English as a Second DialectAn exploratory study on dialect density estimation for children and adult's African American EnglishAfrican American Vernacular English as a Literary Dialect