Feature Request: Add FunASR as Speech-to-Text provider

[LauraGPT]

Feature Request

LibreChat already supports STT — FunASR would be a great addition as a self-hosted, high-performance STT provider.

Why FunASR?

• OpenAI-compatible API: funasr-server provides /v1/audio/transcriptions — may work as drop-in via custom endpoint
• SenseVoice: Ultra-fast non-autoregressive ASR (234M params), 5-10x faster than Whisper, 50+ languages
• Self-hosted & private: No API key, no cloud dependency
• Emotion detection: SenseVoice detects speech emotions (happy/sad/angry/neutral)
• Built-in VAD + punctuation: Complete pipeline in one call

Server setup:

pip install funasr vllm fastapi uvicorn python-multipart
funasr-server --device cuda
# OpenAI-compatible /v1/audio/transcriptions at localhost:8000

• GitHub: https://github.com/modelscope/FunASR (16K+ stars)
• SenseVoice: https://github.com/FunAudioLLM/SenseVoice (8K+ stars)