Whisper

by OpenAI

About the model:

Model for automatic speech recognition (ASR) and speech translation by OpenAI

Model type:

Text

Model performance class:

Latest

Link to the model:

https://huggingface.co/openai/whisper-large-v3-turbo

Base models:

Whisper-large-v3

End model:

Whisper-large-v3-turbo

End model license:

MIT License

About the organisation:

American AI research organisation founded in 2015, widely known for ChatGPT.

Link to the organisation:

https://openai.com

Model release date:

September 2024

Availability

Base Model Data

Specific data sources are not fully documented or available

End User Model Data

Specific data sources are not fully documented or available

Base Model Weights

https://huggingface.co/openai/whisper-large-v3/tree/main

End User Model Weights

https://huggingface.co/openai/whisper-large-v3-turbo/tree/main

Training Code

Inference code is released, but training code is not disclosed.

https://github.com/openai/whisper

Documentation

Code Documentation

The repository includes comprehensive documentation on model usage and inference.

https://github.com/openai/whisper

Hardware Architecture

Architecture details present in the pre-print

https://cdn.openai.com/papers/whisper.pdf

Preprint

https://cdn.openai.com/papers/whisper.pdf

Paper

Paper accepted (and publsihed) in ICML 2023

https://openreview.net/forum?id=Xr12kpEP3G

Modelcard

https://huggingface.co/openai/whisper-large-v3-turbo

Datasheet

Access

Licenses

MIT License, an OSI-approved license

https://huggingface.co/openai/whisper-large-v3-turbo

Is this information not up to date?

Contribute here ->

Supported by the Centre for Language Studies and the Dutch Research Council. Website design & development © 2024 by BSTN. This version of the index generated 09 April 2026, website content last updated 11 March 2026.