Key Info

ElevenLabs officially launched on OpenRouter, offering 9 text-to-speech models covering 90+ languages and 2 speech-to-text models with speaker labels and word-level timestamps, all discounted 50% through October 19, 8am PT.

Highlights

  • Eleven v4 is the most expressive voice model, supporting audio tags like [whispering]
  • v4 Turbo preserves v4 quality for real-time agents
  • Flash v2.5 targets ultra-low latency use cases
  • Scribe v2 Medical is tuned for clinical conversations
  • Use cases include voice agents, customer service, narration, captions, and meeting notes