Hi team,
First of all, I truly appreciate the effort you’ve put into open-sourcing Feros—this is incredibly valuable for builders like me.
I’m very interested in using Feros for production use, but the main limitation for my use case is the lack of strong real-time speech-to-speech models. I prefer not to rely on OpenAI Realtime due to cost concerns.
It would be a game-changer if you could support:
Gemini Live (already on roadmap 👍)
AWS Nova 2 Sonic (request to add)
If these integrations are implemented, I would fully commit to building on Feros.
Additionally, it would be helpful to have:
Clear documentation on VPS requirements (beyond basic specs)
Guidance for scaling in production environments
Thanks again for your work—excited to contribute and build with the community.
Best,
Dharmesh
Hi team,
First of all, I truly appreciate the effort you’ve put into open-sourcing Feros—this is incredibly valuable for builders like me.
I’m very interested in using Feros for production use, but the main limitation for my use case is the lack of strong real-time speech-to-speech models. I prefer not to rely on OpenAI Realtime due to cost concerns.
It would be a game-changer if you could support:
Gemini Live (already on roadmap 👍)
AWS Nova 2 Sonic (request to add)
If these integrations are implemented, I would fully commit to building on Feros.
Additionally, it would be helpful to have:
Clear documentation on VPS requirements (beyond basic specs)
Guidance for scaling in production environments
Thanks again for your work—excited to contribute and build with the community.
Best,
Dharmesh