I’ve always been a fan of the openai/openai-fm repository, but noticed that the barrier to entry (CLI, env setup, API keys) was high for many. I built OpenAI FM to be a "ready-to-use" web version—no deployment required.
Current Status & Business Model: The project is currently in active development. Since running high-quality AI inference at scale incurs significant server and API costs, I’ve implemented a small paywall to keep the project sustainable as an indie developer.
However, I want everyone to be able to test the quality, so there is a free trial available for all users.
Roadmap & Invitation:
Coming Soon: I’m currently building a Text-to-Video pipeline to turn these voices into synchronized video content.
We need your feedback: Since we are in the early stages, I’d love to hear what features you need. More voices? API access? Specific export formats? Please leave your suggestions here, and I will actively work on integrating them.
I'm especially interested in feedback on the mobile UX and the audio latency. Looking forward to your thoughts!