React FastAPI Rasa Integration

I already have the front end in React, a FastAPI back end, Faster-Whisper for speech-to-text, and a Rasa NLU stack—all roughly 95 % complete. What’s missing is the last mile that ties everything together into one seamless query loop.

Scope of work
1. Capture a WAV file from my existing React component, pass it to Faster-Whisper, and return the transcribed text.
2. Hand that text (or raw text input) to Rasa for context handling.
3. Trigger a unified search module, then store both the user query and the generated response in PostgreSQL (the FastAPI ↔ Postgres link still needs to be wired up).
4. Feed Rasa’s final answer back to React, supporting both plain text and TTS output.

Key details
• Input / output modes: STT and text in, TTS and text out.
• Data to persist: user queries and responses only.
• Tech already in place: React, FastAPI, Rasa, Faster-Whisper, PostgreSQL instance (empty).
• Clean, well-commented code and a concise README showing how to spin everything up locally with Docker or simple shell commands.

I’m looking for someone who has actually wired these components together before, can move quickly, and is comfortable debugging across the whole stack. If that sounds like you, let’s finish this loop.