CTranslate2 conversion of OpenAI's Whisper large-v3 turbo for use with faster-whisper and CTranslate2-based projects.
dropbox-dash / faster-whisper-large-v3-turbo is a Hugging Face repository hosting a CTranslate2 conversion of the openai/whisper-large-v3-turbo model. It is intended for developers working with CTranslate2 or projects built on CTranslate2, such as faster-whisper. The model can be loaded in CTranslate2-based pipelines, and it is called automatically for the Mobius Labs fork of faster-whisper.
Weights are saved in FP16. The compute type can be changed when loading the model using the compute_type option in CTranslate2. The listing includes conversion details and points to the original model card for more information. The repository is published under the Dropbox Inc organization.
It is a Hugging Face repository containing a CTranslate2 conversion of the openai/whisper-large-v3-turbo model.
It can be used in CTranslate2 or projects based on CTranslate2 models, such as faster-whisper. It is called automatically for the Mobius Labs fork of faster-whisper.
Yes. The model weights are saved in FP16, and the type can be changed when the model is loaded using the compute_type option in CTranslate2.
The repository points to the original openai/whisper-large-v3-turbo model card for more information.
