litellm - 💡(How to fix) Fix Feature Request: Add FunASR as audio/transcriptions provider

StepCodex · 2026-05-31T02:07:03Z

[litellm] Feature Request LiteLLM already supports /audio/transcriptions with multiple providers. FunASR would be an excellent self-hosted provider addition. W… ## Feature Request LiteLLM already supports `/audio/transcriptions` with multiple providers. FunASR would be an excellent self-hosted provider addition. **Why FunASR?** - **OpenAI-compatible API**: `funasr-server` provides `/v1/audio/transcriptions` — should be straightforward to add as a provider - **SenseVoice**: 50+ languages, 5-10x faster than Whisper, 234M params - **Self-hosted**: No API key, fully local — many LiteLLM users need on-premise STT - **Fun-ASR-Nano**: LLM-based ASR with hotword customization - **Extra capabilities**: Emotion detection, speaker diarization, audio events **FunASR server:** ```bash pip install funasr vllm fastapi uvicorn python-multipart funasr-server --device cuda # /v1/audio/transcriptions at localhost:8000 ``` This could be added as a provider similar to how Whisper/Deepgram/Azure are supported. Users would configure it via: ```yaml model_list: - model_name: funasr litellm_params: model: funasr/sensevoice api_base: http://localhost:8000 ``` - GitHub: https://github.com/modelscope/FunASR (16K+ stars) - SenseVoice: https://github.com/FunAudioLLM/SenseVoice (8K+ stars)

litellm2026-05-31 02:07:03

ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

GitHub issue URL

Helpful · Quick feedback

Code Example

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

---

model_list:
  - model_name: funasr
    litellm_params:
      model: funasr/sensevoice
      api_base: http://localhost:8000

RAW_BUFFERClick to expand / collapse

Feature Request

LiteLLM already supports /audio/transcriptions with multiple providers. FunASR would be an excellent self-hosted provider addition.

Why FunASR?

OpenAI-compatible API: funasr-server provides /v1/audio/transcriptions — should be straightforward to add as a provider
SenseVoice: 50+ languages, 5-10x faster than Whisper, 234M params
Self-hosted: No API key, fully local — many LiteLLM users need on-premise STT
Fun-ASR-Nano: LLM-based ASR with hotword customization
Extra capabilities: Emotion detection, speaker diarization, audio events

FunASR server:

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

This could be added as a provider similar to how Whisper/Deepgram/Azure are supported. Users would configure it via:

model_list:
  - model_name: funasr
    litellm_params:
      model: funasr/sensevoice
      api_base: http://localhost:8000

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

Vote matrix · Quick signals

Works

Did the solution work? Tap to confirm.

Easy Fix

Was it a quick fix?

Time Saver

Did it save you time?

Blocking

Was it severely blocking?

Common Issue

Are others likely hitting this too?

Flaky / Intermittent

Is it intermittent?

Verified / Reproducible

Can you reproduce it reliably?

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Data

Security

Network

Code

UI/UX

Text

System

Multimedia

Protocol

API

Engineering