mcp.film / Voice & Dialogue / ElevenLabs MCP

ElevenLabs MCP

OfficialRemoteby ElevenLabsfreemiumverified Sep 17, 2026

Official ElevenLabs server: TTS, voice cloning, SFX, music, and dialogue isolation.

What it does

The official ElevenLabs MCP exposes the full audio platform: text-to-speech with thousands of voices, voice cloning and design, text-to-sound-effects, Eleven Music composition, transcription, and audio isolation, plus video_to_music to score an existing cut and inpainting to replace a section of a generated track. For a filmmaker-agent it covers dialogue, SFX, and score from one connection — the most complete one-stop audio MCP for a film pipeline. ElevenLabs archived the local stdio server in September 2026 and now ships this as a hosted OAuth endpoint.

Connect

Remote endpoint (Streamable HTTP)

https://api.elevenlabs.io/v1/mcp

Claude Code

claude mcp add --transport http elevenlabs https://api.elevenlabs.io/v1/mcp

Claude Desktop (claude_desktop_config.json)

{
  "mcpServers": {
    "elevenlabs": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://api.elevenlabs.io/v1/mcp"
      ]
    }
  }
}

Cursor (.cursor/mcp.json)

{
  "mcpServers": {
    "elevenlabs": {
      "url": "https://api.elevenlabs.io/v1/mcp"
    }
  }
}

Local install: uvx elevenlabs-mcp (archived local server; needs ELEVENLABS_API_KEY in the environment)

Auth: oauth — Browser OAuth against your ElevenLabs account — no key to rotate; the archived local server still takes ELEVENLABS_API_KEY from the elevenlabs.io dashboard

Sample tools

text_to_speechspeech_to_texttext_to_sound_effectsvoice_clonetext_to_voiceisolate_audiocompose_musiccreate_composition_planvideo_to_musicupload_music_for_inpaintingsearch_voice_libraryplay_audio

Field notes

This entry flipped to the hosted endpoint in September 2026, and the reason is blunt: ElevenLabs archived the local server. github.com/elevenlabs/elevenlabs-mcp is read-only as of 16 Sep 2026 (1.5k stars, last push 20 Aug, MIT) — PyPI elevenlabs-mcp 0.12.2 from 4 Aug 2026 still installs and still runs, so the stdio command above is not broken, but it is frozen and will not get new models. Two days before archiving it, on 14 Sep 2026, ElevenLabs extended the OAuth endpoint to ElevenCreative: speech, Scribe transcription across 99 languages, dubbing, music, sound effects, and 50+ image and video models, in Claude, ChatGPT and Cursor. Treat that as rolling out rather than finished, because the vendor contradicts itself — the hosted-MCP reference docs still describe agent management (create, duplicate, inspect, cost agents) plus a single text-to-speech tool, and elevenlabs.io/mcp still prints 'creative tools coming soon' three days after the announcement. Run tools/list against your own workspace before you plan a pipeline on it. The tools_sample above was verified against the local server, not the remote. Hosted output lands in an ElevenCreative workspace instead of coming back as local files, which is a real pipeline difference if you were shelling out to ffmpeg next. Regional endpoints exist (api.us, plus EU, India and Singapore variants) and workspace admins can restrict tool access and data residency. Credits are consumed by every tool either way. On the archived local path: ELEVENLABS_MCP_BASE_PATH (default ~/Desktop) is a hard security boundary for input files as well as output — any local path a tool reads must resolve inside it, so point it at your media root or reads will be rejected; other env is ELEVENLABS_MCP_OUTPUT_MODE (files/resources/both, default files) and ELEVENLABS_API_RESIDENCY (enterprise-only).

Also in Voice & Dialogue

AudioPod MCPAudioPod AI Official

Hosted MCP for TTS, voice cloning, music, stem and speaker separation, and denoise.

tts · voice-cloningremote · freemium

AllVoiceLab MCPAll Voice Lab Official

Official voice and video localization MCP: TTS, cloning, dubbing, subtitles.

tts · speech-to-speechcredits

Cartesia MCPCartesia Official

Official Cartesia (Sonic) server: low-latency TTS, cloning, voice change, localization.

tts · sttremote · freemium

Hume MCP ServerHume AI Official

Official Hume server for Octave TTS — emotionally expressive, acting-directed speech.

tts · voice-designfreemium

Fish Audio MCPFish Audio Official

Hosted Fish Audio server: voice search, expressive TTS, transcription, plus image and video generation.

tts · voice-libraryremote · freemium

Used in

AI Commercial Sprint · Local Edit Bay · Character-Consistent Series · Make a polished commercial fast · Create avatar, talking-head, or UGC-style video · Generate voice, score, and sound assets

Pairs well with

Artyfile Music LicensingArtyfile LTD Official

Search a catalogue of real recordings and buy a lifetime worldwide sync licence per track, WAV plus certificate.

music-search · music-licensingremote · paid

AudioLab MCPAudioLaunch Official

Loudness, true-peak and voice metering that grades a mix against real delivery specs.

loudness-analysis · true-peakremote · freemium

Epidemic Sound MCP ServerEpidemic Sound Official

Search a cleared music and SFX catalogue, then download stems, from the agent.

music-search · music-licensingremote · paid