Eleven v3 Conversational

Realtime expressive speech for agents and interactive characters (~280ms model latency).

Eleven v3 Conversational targets realtime speech with richer delivery than Flash, at roughly 280ms model latency on the docs. It fits support agents and interactive characters that still need emotion.

Model spelling matches the official docs. For quotas, languages, and the latest changelog, check the official ElevenLabs website.