Log inSubmit server

Hidden while we check this listing

Our automated checks found a problem, so this listing is out of search, browse and the API until it clears. It comes back on its own once the problem is fixed.

  • The install package comes from a different repo than this listing's
Own this project? Claim it
FunASR logo

FunASR

Transcribes local audio with FunASR and SenseVoice through an MCP interface for Claude and Cursor.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install FunASR

Pick your client, then add it in one click or paste the config.

STDIOPython
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "funasr": {
      "command": "uvx",
      "args": [
        "torch"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

We couldn’t confirm it starts correctly

We ran the install command but it didn't respond within our test window — this can mean a slow first-time install rather than a real problem.

uvx torch

Process exited before answering initialize.

This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked 5d ago.

Overview

FunASR MCP server brings local audio transcription from the FunASR toolkit and SenseVoice models to MCP clients. It is intended for private, on-device inference rather than sending recordings to a hosted speech service. Reach for it when an agent needs to process local recordings and you want to use FunASR or SenseVoice capabilities from Claude or Cursor.

Use cases

  • Transcribe local recordings from an MCP client
  • Run private on-device speech recognition
  • Process audio with SenseVoice models
  • Add local transcription to Claude or Cursor workflows

Key features

  • Local audio transcription
  • FunASR model support
  • SenseVoice transcription
  • CPU and GPU inference paths
  • Claude and Cursor compatibility

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked Today

Recent health check succeeded.

11/11 checks healthy over the last 39d

Supply-chain signal

4 high-severity advisories on record for this package. Most advisories affect transitive dependencies and may not be exploitable in this server's actual usage — this is a directional signal, not a security audit.

Critical 2High 2Medium 3Low 36

Scanned 12d ago via OSV.dev ¡ torch (PyPI)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars20kStargazers on the source repository.
Last commitTodayMost recent push to the default branch.
Install checkInconclusiveDidn't respond in our test window — often a slow first install.
Directory activity1 viewsConfig copies, upvotes, and views on AllMCPs.

How FunASR works

What the FunASR MCP server does

The FunASR MCP server connects an MCP-compatible client to local speech-recognition capabilities from the FunASR project. Its documented use case is transcribing local audio with FunASR and SenseVoice while keeping inference on the device. The repository identifies an MCP server example for use with Claude and Cursor.

This fits workflows where an agent needs to work with recordings available on the host machine. Instead, it provides an MCP-facing route to speech models that are selected and executed locally through the FunASR ecosystem.

How it works

FunASR is organized as a toolkit with separate models and runtime paths. The repository documents Python AutoModel usage for models such as Fun-ASR-Nano and SenseVoiceSmall, with CPU and GPU execution depending on the installed PyTorch environment. SenseVoiceSmall is also shown with optional VAD and CAM++ speaker-processing components.

The FunASR MCP server is located in the repository’s examples/mcp_server/ area. At the model level, audio is supplied to a local inference pipeline and the resulting transcription is returned to the calling workflow.

Model selection matters. Fun-ASR-Nano covers Chinese, English, Japanese, and specified Chinese dialect groups and regional accents in the documented base checkpoint. Fun-ASR-MLT-Nano is a separate 31-language checkpoint. SenseVoiceSmall is a five-language checkpoint with emotion and audio-event tags. These capabilities should not be assumed to apply equally to every checkpoint or serving backend.

Tools and capabilities

The documented scope supports these uses:

  • Transcribe local audio with FunASR models.
  • Transcribe audio with SenseVoiceSmall.
  • Run inference on CPU or compatible NVIDIA GPU environments.
  • Process audio through an MCP client such as Claude or Cursor.
  • Use VAD and speaker-aware pipelines when configuring the underlying FunASR components separately.

Speaker processing has a specific boundary: CAM++ embeddings are clustered into anonymous speaker indices that are local to one recording. They do not identify enrolled or known people.

Limitations and notes

Use the repository’s MCP example for those operational details rather than inferring them from the Python quickstarts.

Checkpoint and runtime compatibility are not interchangeable. GPU execution requires a matching PyTorch and torchaudio installation, and the repository advises using CUDA only when the environment detects a compatible GPU. GGUF models use a llama.cpp path and are not Python AutoModel checkpoints. Timestamp, language, speaker, and tagging behavior depends on the selected model and pipeline. Benchmarking with representative audio is recommended before committing to a model or runtime.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about FunASR

FunASR is an MCP server that connects Claude or Cursor to local FunASR and SenseVoice speech-recognition models. Its main tools support audio transcription through on-device inference, allowing recordings to be processed without a hosted transcription service. The exact MCP tool names and transport settings are defined in the repository example.

It points to the repository’s `examples/mcp_server/` directory, which should be used for the server’s setup and client configuration. The general toolkit quickstart installs `torch`, `torchaudio`, and `funasr`, but that is not a complete MCP installation procedure.

Yes, the README explicitly lists the MCP server for Cursor.

Yes, the README explicitly lists Claude among the MCP server clients.

Related MCP servers

All in Speech-to-Text All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewFunASR AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/funasr?style=directory)](https://allmcps.com/mcp/funasr)
HTML Embed
<a href="https://allmcps.com/mcp/funasr"><img src="https://allmcps.com/api/badge/funasr?style=directory" alt="FunASR on AllMCPs" /></a>