ElevenLabs

Voice AI API for text-to-speech, speech-to-text (Scribe), voice cloning, dubbing, music and conversational voice agents. Official SDKs for Python and TypeScript, plus a REST API.

Website Docs Changelog Compare with others

ElevenLabs at a glance

From official pages, checked Nov 15, 2026. Tap the arrow for each source.

Free tier: Yes
Free plan with 10k credits per month (TTS, STT, sound effects and more).
Starting price: $6 per month
Starter plan, 30k credits per month with commercial license.
Rate limit: 2 concurrent requests
Free plan, Multilingual v2 (Flash models: 4 concurrent).
Size limit: 10,000 characters per request
eleven_v4 and Multilingual v2; Flash v2.5 allows 40,000.
Official SDKs: PythonJavaScript / TypeScript
Trains on your data: Opt-out
Data used to improve models by default; toggle off in Data use settings. Enterprise excluded by default.
EU hosting: Yes
Isolated EU data residency environment, Enterprise customers only.
Compliance: SOC 2GDPRHIPAA

What you can build with ElevenLabs

ElevenLabsPro archiveEmbedPaste

Add a talking AI voice agent to any website with the ElevenLabs widget

Paste ElevenLabs' official embed snippet into your HTML and visitors can talk to your agent by voice, or by text if you switch it on.

Free tier: YesFrom $6 per month

AgentsChat

Recent updatesAll updates

Nov 12, 2026
ElevenLabs releases Scribe v3 speech to text and gives sound effects their own concurrency limits (November 9, 2026)
New scribe_v3 and scribe_v3_live models. Sound effects now return 429 past 4 to 8 concurrent requests. audio/pcma is removed.
Nov 7, 2026
ElevenLabs sound effects v3 and agent LLM deprecations (November 2, 2026)
New eleven_text_to_sound_v3 model, v2 removed March 1, 2027. Agents drop gpt-4.1-nano, gpt-5 models and Sonnet 4.5.
Oct 28, 2026
ElevenLabs moves Speech to Speech to Voice Changer paths and adds clone strength (October 2026)
ElevenLabs deprecates /v1/speech-to-speech for /v1/voice-changer, adds clone_strength, callback_disconnect and a fast agent LLM tier.
Oct 23, 2026
ElevenLabs adds Eleven Realtime models for voice agents: v1 Mini and v1 Max (October 19, 2026)
Two managed voice-agent models for ElevenAgents, with OpenAI Realtime API compatibility. In every workspace since November 2.

ElevenLabs at a glance

What you can build with ElevenLabs

Add a talking AI voice agent to any website with the ElevenLabs widget

Generate speech from text with ElevenLabs in Node.js

Recent updatesAll updates