A chapter costs the same as a sentence.
2,000
characters a request, a book is a loop over them
One narrator voice that holds the same register from the first page to the last, cloned from ten seconds, with the pacing and the pronunciation set per request. A book is a few hundred thousand characters by definition, and length is the entire product.
The arithmetic of length
A single request carries up to 2,000 characters, so a chapter is a loop and a book is a longer loop. A 400-page book and a 40-page one cost the same to render.
audio
an opening paragraph, read straight through
generated 2026-08-05
English
“The lighthouse keeper counted ships the way other people counted sheep, and by the end of most winters he could tell a trawler from a tanker by the sound of it alone, without ever leaving his chair. He had kept the light for nineteen years. He had spoken to perhaps forty people in that time, and remembered all of them. On the night this story begins, the sea was flat, the lamp was turning, and one of the ships was counting back.”
Leo · 22.1 s · audiobook · long-form narration
The controls a long read needs
- temperature sets melodic range across its span, so the read can sit anywhere between recited and performed, and a held seed makes the take you liked repeatable.
- cfg_weight loosens the pacing, so a reflective passage gets room instead of being hurried.
- A pronunciation dictionary fixes proper nouns once per request, character names, places, invented words, rather than after a listener complains.
Notes
How do I keep a voice consistent across a whole book?
Every chapter uses the same reference clip, so there is no drift between sessions. A seed is accepted and forwarded, though takes are currently non-deterministic.
What happens to the bill on a very long book?
Nothing. Characters and minutes are unmetered on a stream, so length changes how long the render takes and not what it costs.
A key and one stream to build this on, the same production API this page measures.
Get a key for this use case