Sayak0504
/

whisper-base-bn-roman-vistaar

Model card Files Files and versions

Whisper Base Bengali Romanized

This model is a fine-tuned version of openai/whisper-base on the Bengali subset of the AI4Bharat Kathbath dataset. It transcribes Bengali audio into Romanized (ITRANS) script.

Downloads last month: -

Safetensors

Model size

72.6M params

Tensor type

F32

·

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support