Fun ASR MCP Server

Transcribe local audio with FunASR and SenseVoice using private, on-device inference.

People who work with pytorch, speech recognition and paraformer and want it reachable from Claude, Cursor, VS Code, or another MCP client. The project is written in Python.

VERIFIED ACTIVE

LAST COMMIT 2026-09-18 · ★ 20,411 · #8 OF 106 MAINTAINED AI & ML · VERIFIED 2026-09-18

MIT · Python servers · how we verify → /methodology

01 · Install Fun ASR

Docker

docker run -i --rm ghcr.io/modelscope/funasr-mcp:0.1.2

Claude Desktop - add to config

{
  "mcpServers": {
    "modelscope-funasr": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/modelscope/funasr-mcp:0.1.2"
      ]
    }
  }
}

Using another client? Same JSON, different key

Claude Desktop · mcpServers

Cursor · mcpServers

VS Code · servers

Windsurf · mcpServers

Zed · context_servers

Cline · mcpServers

Roo Code · mcpServers

Continue · mcpServers

LibreChat · mcpServers

Gemini CLI · mcpServers

Codex CLI · mcp_servers

Full setup guides: every client.

02 · Evidence

Security posture

What to check before giving this server access to your agent - from the registry, GitHub, and our own probes. We don't score safety; we show what's verifiable.

runs as container (OCI image) - can be sandboxed away from your filesystem

license MIT - declared in the repository

registry namespace io.github.modelscope is GitHub-verified and matches the repo owner

03 · What Fun ASR can do

Prose above is summarized from the project's README and registry record - no invented capabilities.

Latest releases

v1.4.16 · 2026-09-18

FunASR 1.4.16 · Highlights · Reliable mixed-device pipelines: explicit vad_kwargs, punc_kwargs, and spk_kwargs device placement is preserved across repeated inference calls. Runtime device overrides no longer mutate…

funasr-onnx-v0.4.3 · 2026-09-09

Standalone ONNX Runtime Python package · Install with: · bash · python -m pip install --upgrade "funasr-onnx==0.4.3" · This is the funasr-onnx distribution, not a new release of the separate funasr toolkit. Updating…

v1.4.15 · 2026-09-09

FunASR 1.4.15 · Highlights · NumPy 2 compatibility: remove the NumPy upper bound with real frontend, masking and speaker-clustering tests; update EEND mel-filter calls for modern librosa. Thanks to @linhongyu510 for…

04 · Who maintains Fun ASR

FunASR is maintained by modelscope. It's the only MCP server we track from this author; the repo dates to Nov 2022.

05 · Facts

repository: github.com/modelscope/funasr
website: https://github.com/modelscope/FunASR/tree/main/examples/mcp_server
category: AI & ML - ranked #8 of 106 actively-maintained AI & ML servers as of 2026-09-18.
release cadence: 10+ releases in the last 90 days (latest 2026-09-18)
registry: io.github.modelscope/funasr-mcp (active, first published 2026-07-23 · 2 versions)
packages: oci:ghcr.io/modelscope/funasr-mcp:0.1.2

06 · Fun ASR FAQ

What is Fun ASR?

Transcribe local audio with FunASR and SenseVoice using private, on-device inference.

Is Fun ASR still maintained?

Yes - as of 2026-09-18, its last commit was 2026-09-18 and it shipped 10+ releases in the last 90 days. We re-verify nightly.

How do I install Fun ASR?

Run `docker run -i --rm ghcr.io/modelscope/funasr-mcp:0.1.2`. You can also paste the ready-made client config above.

07 · Alternatives to Fun ASR

Alternatives to Fun ASR

Maintained AI & ML servers if Fun ASR isn't the fit.

Pairs well with

Servers that cover what Fun ASR doesn't - only shown when the pairing reason fits the companion.

More AI & ML MCP servers · Prompt Protection · Mercury Spec Ops · x402 Universal Payer · three.ws 3D Agent Tools · three.ws Activity

More Python MCP servers · jPOS MCP Server · MolTrust MCP Server · Microcap MCP · MemoCat - Shared AI Memory · Montycat MCP - Shared Memory for AI Agents · see all