Audio
Edit manuscript → narrate each chapter → mix with ambient score
5-column Gab AI Deck recipe for indie audiobook production
Audiobook Production turns a manuscript into a finished audiobook on a parallel column-grid. Column 1 proofreads each chapter for narration quirks (em-dash pauses, name pronunciations, dialogue tags). Voice columns 2–3 narrate sequential chapters; the music column lays an ambient score; the tools column merges stems into chapter masters. Save the deck as a recipe and the next book starts on chapter one of the workflow, not the dread of one.
ElevenLabs produces remarkably consistent narration when you pin a single voice profile across columns. For multi-voice (dialogue, character voices) split chapters into segments and bind each to a different voice profile.
Most TTS columns handle up to ~30 minutes per run cleanly. For longer chapters, split into 5–10 minute scenes and merge in the tools column.
Yes — fade the score out under the spoken word and only let it breathe between chapters. Spec that explicitly in the music column prompt.
The output is broadcast-quality WAV at 192kbps+; for ACX, run a final mastering pass with peak limiter and loudness normalization (LUFS) in your DAW or audio tool of choice.
Yes — ElevenLabs supports multilingual narration. Translate the chapter in a chat column first, then route the translation through narration with the matching voice profile.
Save the deck as a recipe; the next book clone inherits the same voice binding so chapter one of book two sounds like book one.