ElevenLabs · Text to speech

Eleven v4

Eleven v4 is ElevenLabs' most expressive speech model, suited to narration, audiobook excerpts, advertising voiceovers and character lines where emotional delivery matters. It follows inline audio tags in brackets, such as [whispers] or [laughs], to shape a performance, and speaks more than 90 languages fluently. Flowctory generates an MP3 with one selected voice and a stability choice: lower is more expressive, higher is more consistent between takes. Review pronunciation and tone before publishing.

Generation mode

Text to speech

Credit cost

Price depends on settings

Observed generation time

Not enough public data

Release date

Sep 28, 2026

Current credit pricing

The credit cost is calculated from the selected input, output, duration, or quality settings and has no single fixed row.

Availability and credit rows come from Flowctory's active production catalog. Observed time is shown only after at least five completed generations and is not a controlled benchmark. Model availability, limits, and costs can change.

Use model