studio & library/Text to Speech

Text to Speech

A three-pane studio for turning text into speech: browse voices, write the script, tune the delivery.

Narrating a script — The second half of the voices walkthrough: the Text to Speech screen, its engine settings, and generating speech from a pasted script.

The three panes

Voice browser: Every available voice, searchable and filterable, previewable before you commit.
Script: What will be spoken, with a live character count and the cost of the current engine.
Settings: The chosen engine’s own parameters — these differ per engine, because they genuinely differ per engine.

Models

The model list is flat: each of Kokoro, Qwen3 and OmniVoice is one row, and each ElevenLabs model is its own row. Local engines cost nothing per character; ElevenLabs rows show a price per thousand characters.

History

Every synthesis is kept in the History tab, playable and downloadable, so you can compare takes rather than regenerating from memory. Clips also land in the Voiceovers grid of your Library.

Related

Voices Library Tokens & usage

PreviousVoices NextAvatars

Didn’t find your answer?

Ask us directly and we’ll walk you through it — your question usually becomes the next page here.