

Eleven Labs
#11 v Syntéza řeči (TTS)Eleven Labs · od April 2022 (Unternehmensgründung); öffentliche Beta-Plattform Januar 2023 · 2× · naposledy 03. 10. 2026
ElevenLabs is an AI platform for speech synthesis (text-to-speech), voice cloning, speech-to-text, and conversational agents, founded in 2022 by Piotr Dąbkowski and Mati Staniszewski. Its current flagship models (Eleven v4, v4 Turbo, v3) support over 90 languages, offer latencies as low as ~75-100ms, and provide real-time streaming via WebSocket/HTTP. The product is sold through a tiered subscription model (Free to Enterprise, $0-$990/month) plus usage-based API pricing; commercial use is reserved for paying users only, and the underlying software is proprietary (closed source).
Vlastnosti
| Real-Time Streaming | Yes, via REST (Server-Sent Events) and WebSocket API, playback starts before full generation completes |
| Latency | Ultra-low latency from ~75ms (Flash v2.5) resp. ~100ms (Eleven v4 Turbo); 250-300ms for Turbo v2.5 |
| License | Proprietary/closed source; commercial use of audio outputs only permitted with a paid plan |
| Platform | Web app, API (REST & WebSocket), iOS and Android app |
| Price | Free $0; Starter $6/month; Creator $22/month; Pro $99/month; Scale $299/month; Business $990/month; Enterprise custom |
| Release Date | Company founded April 2022, public beta platform launched January 2023 |
| Languages | 29 to 90+ languages depending on model (Eleven v4/v4 Turbo: 90+, Eleven v3: 70+, Multilingual v2: 29) |
| Voice Cloning | Instant Voice Cloning from Starter plan ($6/month); Professional Voice Cloning from Creator plan ($22/month) |