A5100620754- Subtitles and Audiovisual Media
- Music and Audio Processing
- Video Analysis and Summarization
- Second Language Learning and Teaching
- Syntax, Semantics, Linguistic Variation
- Language, Discourse, Communication Strategies
- Linguistic Variation and Morphology
- EFL/ESL Teaching and Learning
University of California, Los Angeles
2021
Hunan University of Technology and Business
2013
Generative models have shown significant achievements in audio generation tasks. However, existing struggle with complex and detailed prompts, leading to potential performance degradation. We hypothesize that this problem stems from the low quality relatively small quantity of training data. In work, we aim create a large-scale dataset rich captions for improving models. develop an automated pipeline generate audio-visual datasets by transforming predicted visual captions, tagging labels...
Focusing on the Foto dialect of Dschang (Yemba), an understudied Grassfields Bantu language spoken in Cameroon, this paper offers a cross-linguistic perspective Cognate Objects (CO). An argument analysis COs is supported by both comparison, e.g. forms corresponding wh-questions, compatibility with strong determiners, quantifiers and possessors, ability to be pronominalized relativized, Dschang-internal evidence including word order variations tonal marking object position.