XLSR models fine-tuned on 100 hours of English dialects from North America, UK, Argentina, India, Korea, Spain, China, Middle East, Vietnam, and more!
Koel Labs
company
Verified
AI & ML interests
Speech models, phonemic transcription, fine-tuning
Recent Activity
View all activity
XLSR models fine-tuned on 100 hours of English dialects from North America, UK, Argentina, India, Korea, Spain, China, Middle East, Vietnam, and more!
Existing speech corpora preprocessed into Hugging Face datasets with a consistent format and column naming scheme.
XLSR models fine-tuned on 5 hours of North American English dialects with various augmentation strategies. Tiny vocab of 40 most common sounds.
New datasets for IPA transcription.