From a single prompt describing the idea, Narrate drafts a narrated audio edition — voice performance, pacing and chaptering — then applies audio mastering so the draft is technically finished. Given an optional source manuscript, it narrates that existing text instead of originating new material. Trained on narrated long-form, distinct from the dramatized, multi-voice work Drama performs. It is an operational reference, not an offer or investment content.
Narrate is not a text-to-speech wrapper with a chapter marker bolted on. It is trained specifically on narrated long-form audio — the pacing, chaptering and mastering a finished audiobook needs — so an audio draft comes back as a finished audiobook, not a raw read-through that still needs production.
Narrate returns a finished audiobook draft in the state a retail or library channel expects — narrated, paced, chaptered and mastered — rather than a raw voice file that still needs an audio engineer.
The draft is narrated with the pacing, emphasis and continuity a long-form audiobook requires — a performed read, not a flat text-to-speech output. The voice carries the work across chapters.
Chapter breaks, pacing and continuity are drafted with the structure an audiobook needs — the listener can navigate the work chapter by chapter, rather than receiving a single unbroken audio file.
Mastering is applied to the draft so the audio is technically finished — consistent levels and the technical standard a retail or library channel expects — rather than a raw file needing a post-production pass.
If you have an existing manuscript you would rather narrate than originate, upload it alongside the prompt. Narrate returns the same finished audio draft, narrated from that source rather than from a fresh draft.
Narrate is trained specifically on narrated long-form audio, distinct from the dramatized, multi-voice work the Drama model performs. See usage & pricing for per-model unit costs.
Narrate follows the same single-prompt pattern as every model built on PUBXX 1.0: one instruction goes in — describing an idea, or naming an optional uploaded source to narrate — the engine drafts, and Narrate shapes that draft into a narrated, mastered audiobook. There is no multi-step chain to manage, and no required human pass before a draft is returned.
By default, a single prompt describing the idea is the only input. To narrate an existing work instead, a source manuscript can be uploaded alongside it.
The core engine produces the structural and language-consistent base draft that Narrate then shapes into the audiobook format.
Narrate applies narration, pacing, chaptering and audio mastering, and returns a complete audiobook draft.
Editorial review is available as a separate, optional step after a draft is returned, priced on its own — it is not part of the single-prompt generation described above. See usage & pricing for details.
Narrate does not solve pacing, continuity, formatting logic and metadata from nothing. It inherits those from PUBXX 1.0, the shared drafting and consistency layer that every named model is built on — and specialises that layer for the audiobook format.
PUBXX 1.0 handles the parts of production that are skilled but repetitive: structural and language consistency, formatting logic and metadata handling. Narrate is what that engine becomes when specialised for the audiobook — a format-specific model, built on a shared core, not a separate, unrelated tool.
Narrate is one of five named models sharing the same PUBXX 1.0 engine. See the full engine breakdown for how the others work.
Narrate is built to produce drafts at a consistency that hand assembly across separate tools cannot sustain — a consistency the shared PUBXX 1.0 engine is designed to keep, rather than a property of any single model in isolation.
Human audiobook narration runs $200–$600 per finished hour, or $2,500–$4,900 for a full novel. Narrate is built to reduce that line. Cost reduction is company-reported, against an all-human workflow, and is not a revenue claim.
Narrate is the natural model for any creator who needs a finished audiobook edition — whether originating from scratch or narrating an existing text. It is not tied to any single platform or token; it is the infrastructure underneath SEW Capital's audiobook output generally.
SEW Capital owns 100% of PUBXX 1.0 Narrate — the model, the engine underneath it, and the IP and output the model generates. No token confers any ownership of the model, the engine, or their output.
This page describes the Narrate model as infrastructure in its own right. It does not describe, and should not be read as describing, the utility of any specific token or the mechanics of any specific platform that may use this model's output.
This page describes PUBXX 1.0 Narrate, the audiobook model built on SEW Capital's AI engine. It is an operational reference. It is not a prospectus, not an offer, not a solicitation, and not a recommendation to acquire, hold or dispose of any token.
Not tied to a platform or token. This page intentionally does not describe how any specific platform sequences its workflow, nor the utility, mechanics or settlement of any specific token. Those are documented separately, elsewhere in the insights list, and token usage pricing for this model is set out on the usage & pricing page.
Not a forecast. The figures on this page describe the model's design and SEW Capital's own catalogue performance to date. They are not evidence of how any individual future title, format or language edition will perform.
We do not publish price information, return expectations, yield, appreciation claims, or any statement of value for any token on this page. Nothing here should be read as financial, legal or tax advice.
We can take you through how the model narrates and masters, what a finished audiobook draft contains, and how it fits the shared engine underneath. This is documentation, not a pitch.
hello@sewcapital.co.uk See usage & pricing