Eleven v4 and Eleven v4 Turbo

Eleven v4 is built on an entirely new architecture that reads a script the way a voice actor would: it knows who is speaking, what just happened and how each line should land. Direction goes straight into the script as tags like [whispers], [nervous laugh] or [door slams], and v4 follows them more reliably than v3, sound effects included, across 90+ languages and multiple speakers in one pass. Eleven v4 Turbo carries the same expressive range at about 150 ms to first speech, for real-time agents. Artificial Analysis ranked it first at launch.

For filmmakers the headline is consistency. Regenerate a line once or fifty times and it is still the same person speaking, with no vocal drift, and context stitching keeps pacing and delivery steady across a script of any length, so a long piece sounds like one take. Professional Voice Clones, missing from v3, are back with the full emotional range. ElevenLabs had voiced AI films since the first Gen-2 shorts; v4 turns it from a line reader into something closer to a cast you can direct.