What it's for

elevenlabs turns a Markdown script into AI narration through the ElevenLabs API, producing one audio clip per paragraph rather than a single monolithic file.

The point is the content-addressed cache: each paragraph's audio is keyed by its own text and voice settings, not by its position in the script. Edit one paragraph — or split it in two — and every other clip is served from cache. On a script you revise five or six times before recording, that is the difference between paying for the whole narration each pass and paying only for the sentences you actually changed.

Per-paragraph clips also make assembly easier: each one lines up with a section of the edit instead of needing to be cut out of a continuous take.

Install

dotnet tool install --global grdev.elevenlabs-cli

Published on NuGet as grdev.elevenlabs-cli.

Requires an ElevenLabs API key.

Source

github.com/greek-developer/elevenlabs-cli — currently a private repository.