Omar Sanseviero

Actually I misread the code. Now I realize this is a ASR Model, for which we don't have support in `fairseq` in the API. I'll add `asr`, but the model...

We do allow passing `top_k` and `top_p` for generation tasks at the moment (see https://github.com/huggingface/huggingface_hub/blob/main/api-inference-community/api_inference_community/validation.py#L81-L82. although this seems unused). But this is passed in the input, not as parameters :thinking:....

cc @sanchit-gandhi @Vaibhavs10

cc @Vaibhavs10 @sanchit-gandhi

Hi there! 🤗 Not at the moment. I would suggest to ask in https://discuss.huggingface.co/ and you can link to the corresponding blog post,

cc @pcuenca and @philschmid as well here > If we need to control the length of input sequences should we initialize tokenizer with model_max_length=X, truncation=True? Yes. > Shouldn't we then...

cc @sanchit-gandhi @Vaibhavs10

I think we can aim to release on Monday. WDYT?

cc @simoninithomas

cc @clefourrier