Skip to content

Release Dual-Use AV-ASR checkpoints on Hugging Face #1

@NielsRogge

Description

@NielsRogge

Hi @ifnspaml 🤗

I'm Niels and work as part of the open-source team at Hugging Face. I discovered your work on Arxiv and was wondering whether you would like to submit it to hf.co/papers to improve its discoverability. If you are one of the authors, you can submit it at https://huggingface.co/papers/submit.

The paper page lets people discuss about your paper and lets them find artifacts about it (your models for instance), you can also claim the paper as yours which will show up on your public profile at HF, and add Github and project page URLs.

I saw in your paper that you've established a new state-of-the-art on the LRS3 AV-ASR benchmark using your dual-use visual fusion method with Whisper. Would you like to host the pre-trained checkpoints (specifically the fine-tuned Whisper small and medium variants) on https://huggingface.co/models?

Hosting on Hugging Face will give your work more visibility and enable better discoverability. We can add metadata tags so that researchers looking for noise-robust or audiovisual ASR models can find your work easily, and link the models directly to the paper page.

If you're down, leaving a guide here. Since Whisper is natively supported in our transformers library, people could easily load your models if you choose to integrate them that way. Alternatively, you can use the PyTorchModelHubMixin class or simply use hf_hub_download.

You can also build a demo for your model on Spaces, we can provide you a ZeroGPU grant, which gives you A100 GPUs for free.

Let me know if you're interested or need any guidance :)

Kind regards,

Niels

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions