QQQ

Meta-Voice — Frontend

Meta-Voice — Frontend (with AI sample voices)

Upload your speaker voice or choose one of the built-in AI voices. Then upload a video and get the converted result.


Tip: clean, quiet recording gives best results.
Note: backend must serve sample files at /samples/{name}.mp3 (example: /samples/meta_female_1.mp3).
The server will extract audio, run voice conversion, and return a new video.
Progress
Make sure your backend API (FastAPI) is running and reachable. This frontend expects these endpoints:
  • POST /upload-voice — accepts form field named file (voice WAV) — used only when "Use uploaded voice sample" is chosen.
  • POST /process-video — accepts form field named file (video MP4). If using an AI sample voice, it should also accept a form field named voice with the sample voice key (e.g. meta_female_1).
  • Server should serve sample mp3s at /samples/{name}.mp3 and outputs at /outputs/.
Example FastAPI static mounts: 
app.mount("/outputs", StaticFiles(directory="outputs"), name="outputs")
app.mount("/samples", StaticFiles(directory="samples"), name="samples")