Spaces:
Running
on
CPU Upgrade
Adding TTS model, using API
Hi and thanks for creating this TTS Leaderboard!
Do you have a fixed template for adding a new model to the Leaderboard?
Let's say I have some API-only TTS model, which I want to evaluate ? Can I add it? Or should it be only an open-source one ?
Looking forward to your reply!
Thanks in advance!
TTS-Arena seems to contact a private Repo:
https://huggingface.co/spaces/TTS-AGI/tts-router
So there is no way for us to help add models with PRs.
Hi, thank you for your interest! If you would like to add a model, please let us know through a Community discussion and we will look into it!
Hi @Pendrokar and @VSydorskyi - We follow the same pattern as LYMSYS chat arena. Where based on community reaction we add models or APIs. In addition to this we do not want to flood the leaderboard with a lot of models in one go so we'll gradually add models one by one over the coming weeks. This allows for a slow but steady increase in the leaderboards' reliability as well.
That said, as @mrfakename mentioned, we encourage users to open a discussion, and then we gauge community reaction and add the models.
Cheers!
VB
I see. Though adding models without a reset seems unfair. But maybe there will be a reset, I am unfamiliar with the LYMSYS inner workings.
@VSydorskyi
I created #30 pull request and you can clone the Space to add the TTS Space you wanted, just add it with parameter overrides if needed right around here:
https://huggingface.co/spaces/Pendrokar/TTS-Spaces-Arena/blob/main/app.py#L54-L65
Hi!
Thanks for all your replies! I have finally prepared API for TTS and will create separate Community discussion