Agents can drive it: the same edit commands the UI uses are exposed over JSON-RPC, gRPC and MCP, so an agent can open a project, cut it and export it.
The AI features run on your own device: text-to-speech, transcription, voice isolation, background removal and beat detection. Nothing is uploaded.
It started about three months ago. It's AGPL-3.0, and bug reports and PRs are welcome.
calanus•13m ago
Agents can drive it: the same edit commands the UI uses are exposed over JSON-RPC, gRPC and MCP, so an agent can open a project, cut it and export it.
The AI features run on your own device: text-to-speech, transcription, voice isolation, background removal and beat detection. Nothing is uploaded.
It started about three months ago. It's AGPL-3.0, and bug reports and PRs are welcome.