OneASR Gateway Integration

OneASR is our open-source, high-performance speech recognition gateway designed for self-hosting and team deployments.

---

Why Use OneASR with DuRT?

• Multi-Engine Multiplexing: Aggregates Faster-Whisper, FireRedASR, SenseVoice, and Paraformer into a single OpenAI-compatible API.
• Async Batch Tasks: Upload 2GB+ video files; OneASR processes them on GPU worker nodes and notifies DuRT upon completion.
• Native Diarization & VAD: Built-in Silero-VAD chunking and speaker diarization pipelines.

---

Quick Setup

1. Deploy OneASR on your local network or GPU server:

   git clone https://github.com/codeAndxv/OneASR.git
   cd OneASR
   pip install -r requirements.txt
   uvicorn app.main:app --host 0.0.0.0 --port 8000

2. In DuRT, go to Settings ➔ Services ➔ Add Service:
   • Type: OneASR
   • Base URL: http://localhost:8000 (or your server IP)
   • API Key: sk-oneasr-xxx (from your config.yaml)
3. Select OneASR in any File Transcription or Live Caption workflow!