← Back to directory

ElevenLabs

Bringing technology to life

Visit ElevenLabs

ElevenLabs is an AI audio and voice-agent platform: text-to-speech, voice cloning, speech-to-text, dubbing, music and sound-effect generation on the "ElevenCreative" side, plus multilingual conversational voice agents for phone, chat, email, and WhatsApp on the "ElevenAgents" side, all accessible through a shared API/SDK layer.

Vendor's own claim
activation/delivery

Core capabilities

Vendor's own claim
  • Text-to-speech across 70+ languages with 5,000+ stock and cloned voices
  • Instant and professional voice cloning
  • Speech-to-text transcription (Scribe v2), including speaker identification
  • Dubbing that preserves emotional tone across languages
  • Music generation and sound-effects creation from natural-language prompts
  • Multilingual conversational voice agents deployable to phone, chat, email, and WhatsApp
  • Video generation and image editing tools
  • APIs and SDKs for developer integration, plus an official MCP server for use inside Claude Desktop, Cursor, Windsurf, and OpenAI Agents

API & agent surface

From official docs
Authentication
API key, sent as a header (xi-api-key)
Base URL
https://api.elevenlabs.io
Notable endpoints
  • GET /v1/history - list generated audio items (speech history)
  • POST /v1/convai/mcp-servers - register an external MCP server for use by a conversational agent
  • text_to_speech.convert() - SDK-level text-to-speech conversion call shown in the quickstart
Machine-readable spec
OpenAPI spec retrieved

Official skills & repos

From official docs

Published by the vendor themselves — not community forks.

  • elevenlabs/elevenlabs-mcp1494 stars

    Official ElevenLabs MCP server exposing text-to-speech, voice cloning, transcription, voice design, and audio-isolation tools to MCP clients like Claude Desktop, Cursor, Windsurf, and OpenAI Agents.

From the vendor's channel

Sarah's Voice | Creating Art With Her Own VoiceElevenLabs

Pricing

From official docs
PlanPriceIncludedNotable
Free$0/mo10,000 credits/moText to Speech, Speech to Text, Sound Effects
Starter$6/mo30,000 credits/moCommercial license, Instant Voice Cloning, 20 Projects in Studio
Creator$22/mo ($11 first month)121,000 credits/moProfessional Voice Cloning, Higher credit allotment than Starter
Pro$99/mo600,000 credits/mo44.1kHz PCM audio output via API, 192kbps quality audio
Scale$299/mo1.8M credits/mo
3 workspace seats
Team collaboration, 3 professional voice clones
Business$990/mo6M credits/mo
10 workspace seats
Low-latency TTS at 5 cents/minute, 10 professional voice clones
EnterpriseContact salesCustom creditsCustom SSO, DPA/SLA assurance, HIPAA BAAs

Checked 2026-07-30 against their pricing page. Vendors change pricing without notice.

What is ElevenLabs

Vendor's own claim

ElevenLabs began as a text-to-speech/voice-cloning API and has expanded into a broader audio and conversational-AI platform. Its product line splits into ElevenCreative (voice, music, sound-effect, and video generation for content teams) and ElevenAgents (deployable voice agents for customer-facing conversation). Everything runs on a shared credit system and is exposed through REST APIs, official SDKs, and an official MCP server, which is why it shows up heavily in developer and agent-tooling contexts as well as content-production ones.

Ideal customer profile

Vendor's own claim

Developers and content teams building voice/audio features, plus enterprises deploying conversational voice agents; named customers on their site include Twilio, Disney, Epic Games, Nvidia, Salesforce, Revolut, and Meta.

Who should look elsewhere

Our read

The credit-metered pricing (10,000 free credits/month, scaling steeply through paid tiers) makes ElevenLabs expensive for high-volume, always-on narration or long-form audio compared to flat-rate or open-source TTS. Teams needing on-premise or fully self-hosted voice models for data-sovereignty reasons will need to evaluate the Enterprise data-residency options specifically rather than assume the standard API fits, since standard plans are cloud-only.

ElevenLabs alternatives

Our read

Platforms for building and running conversational voice agents that place or answer phone calls.

Category: Voice AI agents. Membership is decided by 2 stated tests, not by category tags — see them and the full comparison.

ElevenLabs FAQ

What does ElevenLabs actually do?

It generates and processes audio with AI: text-to-speech, voice cloning, speech-to-text, dubbing, music/sound-effects, and deployable conversational voice agents, all behind a shared API and credit system.

Vendor's own claim
Is there a free tier?

Yes, a permanent Free plan with 10,000 credits per month covering text-to-speech, speech-to-text, sound effects, voice design, music, and up to 3 Studio projects.

From official docs
Is there an MCP server?

Yes - the official elevenlabs-mcp GitHub repo (about 1,500 stars) exposes ElevenLabs' text-to-speech, voice cloning, transcription, and audio-isolation capabilities to MCP clients including Claude Desktop, Cursor, and Windsurf.

From official docs
How is pricing structured?

A single credit pool shared across all products (TTS, STT, music, video, etc.), metered per plan from Free (10,000 credits) up through Business (6M credits, 10 seats) and custom Enterprise; annual billing is discounted to 10 months' cost, and unused credits roll over up to two months.

From official docs
What company sizes does it fit?

The pricing ladder and named customers (Twilio, Disney, Epic Games, Nvidia, Salesforce, Meta) span solo developers on the free tier through large enterprises on custom SSO/HIPAA plans.

Vendor's own claim
Can it deploy customer-facing voice agents?

Yes - the ElevenAgents line supports multilingual conversational voice agents deployed over phone, chat, email, and WhatsApp, and agents can be wired to external MCP servers via the Conversational AI API.

From official docs
Does it support enterprise compliance needs?

The Enterprise plan lists custom SSO, DPA/SLA assurance, HIPAA BAAs, and data-residency options.

Vendor's own claim
How fast can a developer get a first result?

Signing up and generating a first text-to-speech clip via the web app or a single API call takes minutes; the official MCP server has a documented quickstart for Claude Desktop that also takes only a few minutes.

Our read

Verify before you buy

  • Exact rate limits per plan tier
  • Exact MCP tool names/signatures (the elevenlabs-mcp README documents setup and behavior but does not list tool names in a single enumerable block)

Need help setting it up?

This page tells you whether ElevenLabs fits. Cheetah builds and runs the system it goes into — enrichment, sequencing, CRM and the joins between them.

Book a call →

The directory stays free. There is nothing gated behind this.