How to Connect ElevenLabs to ChatGPT via MCP

💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → ElevenLabs, paste your credentials, mint an MCP API key and you're done. No Docker, no git clone, no local server.

ElevenLabs + ChatGPT

Let your AI use ElevenLabs with your own key: find voices in your account and the Voice Library, turn text into speech you can download, transcribe audio and video, dub videos into other languages, check credits and history, and read your voice agents' conversations.

Prerequisites

See the full setup instructions baked into the connector (visible in the in-app store when you select the connector). The required environment variables for this connector are:

ELEVENLABS_API_KEY

Step 1 — Get credentials

1. Sign in at https://elevenlabs.io and open Settings → API Keys (https://elevenlabs.io/app/settings/api-keys). For a key that does not depend on one person, a workspace admin can create it for a service account under Workspace settings → Service Accounts.
2. Create a key. Leave it unrestricted or allow at least: Text to Speech and Flows (for speech), Speech to Text, Models (read), Voices (read), Add voice from Voice Library, Speech History (read; write only for deletes), User (read), Dubbing (read and write) and ElevenAgents (read). A credit quota on the key caps what it can spend; an IP allowlist would block AnythingMCP.
3. Copy the key (it starts with sk_ and is shown only once), paste it into ELEVENLABS_API_KEY and install the connector. The key is sent only to ElevenLabs.
4. Start with elevenlabs_get_subscription: it shows your plan, the credits used and left, and when they reset.

Data residency: accounts on an isolated environment have their own host and their own keys. Change the connector's base URL to https://api.eu.residency.elevenlabs.io (EU), https://api.in.residency.elevenlabs.io (India) or https://api.sg.residency.elevenlabs.io (Singapore). A key from an isolated environment is refused on the standard host with 401, and the other way round.

Billed to your ElevenLabs account: speech, transcription and dubbing use your credits on a subscription plan, or US dollars on pay-as-you-go API billing. Everything else is free. Approximate prices in October 2026:

• Speech: about 1 credit per character with eleven_multilingual_v2 and eleven_v3, 0.5 with eleven_flash_v2_5 (pay-as-you-go: $0.08 or $0.04 per 1,000 characters). Failed generations are not charged.
• Transcription: about $0.22 per hour of audio.
• Dubbing: about $0.33 a minute with a watermark, $0.50 without.

Confirm first: before speech, the transcription of a long recording or a dub, the AI tells you roughly what it will cost and waits for your yes. Saving a Voice Library voice with elevenlabs_add_library_voice is free but takes one of your voice slots, so the AI asks first too. elevenlabs_delete_history_item is switched off at install.

Speech: elevenlabs_create_speech starts the audio from your text and a voice and returns an id. A few seconds later, elevenlabs_get_speech gives you an MP3 link to listen to or download. The link expires after about an hour; ask again for a fresh one. Speech is made in the background, which works with three models:

• eleven_multilingual_v2 (default): natural and stable, 29 languages, up to 10,000 characters.
• eleven_v3: the most expressive, 70+ languages, audio tags such as [whispers], up to 5,000 characters.
• eleven_flash_v2_5: the fastest and half the price, up to 40,000 characters.

Transcription and dubbing work from a link: you pass a link to the file, not an upload. elevenlabs_transcribe reads a public link (also YouTube or TikTok, files up to 2 GB) and returns the text right away; with speakers separated, elevenlabs_get_transcript shows who said what and when. If ElevenLabs cannot open the link, the connector downloads a direct file link of up to 10 MB and uploads it. elevenlabs_create_dubbing dubs a video or audio file into another language and keeps the speakers' voices; the AI checks elevenlabs_get_dubbing every 30 to 60 seconds until the dub is done. elevenlabs_get_dubbing_transcript returns its subtitles as SRT or WebVTT, and you watch and download the dubbed file in the ElevenLabs app under Dubbing.

What else it reads: your voices and the public Voice Library, filtered by gender, age, accent, language or use case; the models with their limits; the history of generated audio (text, voice, model, date); your plan, credits and voice slots; and your voice agents with their prompt, first message, voice and LLM, plus their conversations with transcript, summary and evaluation. Audio comes back as download links, never as raw audio in the chat; older history items, dubs and call recordings play in the ElevenLabs app.

Common errors: 401 invalid_api_key: the key is wrong, expired or disabled, or belongs to an isolated environment (change the base URL). 400 invalid_api_key: the key was cut off or mistyped. 402 insufficient_credits: no credits left this period, or the key's own quota is used up. 403 insufficient_permissions: the key may not use this endpoint, or an IP allowlist blocks the request; edit the key. 400 or 422: a parameter the model does not accept, for example a language code with eleven_multilingual_v2 or a text above the model's limit. 404: an id that does not exist; the AI checks it with the list tools. 429: too many requests at once; wait a moment and retry once.

The connector does not cover sound effects and music, real-time and streaming audio, voice cloning and voice design, speech to speech, audio isolation, Studio projects, creating or editing agents, phone and batch calls, or downloading the audio of history items, dubs and agent calls.

Step 2 — Install the adapter

curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d

Step 3 — Add the connector in ChatGPT

Your server URL: in AnythingMCP, open MCP Servers → the server this connector is on, and copy its URL (https://cloud.anythingmcp.com/mcp/…). Use it wherever this guide shows YOUR_SERVER_ID.

ChatGPT adds MCP servers as plugins on its Plugins page, on the web and on paid plans. Developer mode is no longer needed. In a Business workspace only admins and owners can add one; in Enterprise and Edu an admin can grant the Create plugins with MCPs permission.

1. Go to chatgpt.com/plugins, click +, then Add custom MCP server.
2. Enter a name, e.g. Name: ElevenLabs.
3. Under Connection, enter the Server URL https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID and keep Authentication on OAuth.
4. Select I understand and want to continue, then Create as a plugin. When ChatGPT asks, sign in to AnythingMCP and allow access.
5. Install the new plugin from your personal plugins, then type @ in a chat and pick it.

Available tools

Example prompts

• "Find three calm British narrator voices in the Voice Library and give me their preview links."
• "Read this product announcement aloud with one of my voices using eleven_multilingual_v2. Tell me the character count and cost first, then give me the MP3 link."
• "Transcribe this YouTube interview with the speakers separated and summarize what each person said."
• "Dub this 4-minute video into German without a watermark. Tell me the cost before you start and let me know when the dub is ready."
• "How many credits do I have left this month, and how did yesterday's calls with our support agent end?"

FAQ

Does the free ChatGPT plan work with custom MCP servers? OpenAI documents custom MCP servers for paid plans only; the Free plan is not on its list.

Next steps

• Connector landing page
• Connect to Claude
• Connect to Cursor
• Connect to Gemini
• Connect to Meta Muse