Voice and music · by ElevenLabs
ElevenLabs
Turn text into lifelike speech, or clone your own voice.
text → speech · text → speech, data · speech → profile
- Way in
- Official APISign up yourself, api key
- To start
- Free tier10k credits/month on Free; commercial licence starts at Starter ($6/mo)
- Runs
- In the cloudAlso through fal, Replicate, OpenRouter, runway, pika
- Checked
- Nov 3, 2026Prices as of that day
What it can do 9
Synthesize expressive speech from text (Eleven v4, 90+ languages, audio tags)10,000 chars/request (v4); 40,000 (Flash v2.5)text → speech$0.08/1k chars (v4); $0.04/1k (v4 Turbo, Flash)
Stream low-latency speech over WebSocket for agents (v4 Turbo / Flash v2.5)text → speech, data$0.04/1k chars
Clone a voice from samples (instant) or train a professional clonespeech → profileincluded in plans (IVC Starter $6/mo+, PVC Creator+)
Design a new voice from a text descriptiontext → profile, speechbilled as TTS characters
Convert one voice recording into another voice (speech-to-speech)speech, profile → speech$0.12/min
Dub audio/video into other languages preserving speaker voicevideo, audio, link → video, speech$0.33/min (watermark) or $0.50/min (v1); $2.20/min (Dubbing v2)
Isolate voice from background noise/musicaudio → speech$0.12/min
Align a transcript to audio (word timestamps)speech, text → datasee pricing page
Dub a video file end-to-end and return dubbed audio plus a preview video (Dubbing v2)API files up to 3 GB, source up to 4K; 3 concurrent jobs self-servevideo, link → video, speechv2 $2.20/min; v1 $0.33/min watermarked, $0.50/min clean
Does the same job
Prices and limits as of Nov 3, 2026, from ElevenLabs’s own pages. They change; check before you rely on them.Not affiliated with ElevenLabs. Not advice.Rankings: Artificial Analysis.Something wrong?