Audio Generation

Generate Tab Overview

The Generate tab in the left sidebar provides controls for generating audio. You can generate audio for individual segments, entire chapters, or the whole project.

Generating Individual Snippets

Click the Zap button (⚡) on any segment card to generate audio for just that segment. This is useful for re-generating a single segment after editing its text or changing its voice.

Generate All

Use Generate All to produce audio for multiple segments at once:

• Chapter level — Generate all segments in the selected chapter.
• Project level — Generate all segments across all chapters.

Progress Tracking

During generation, a progress bar shows how many segments have been completed. Each segment card also shows its individual status.

Snippet Status Indicators

• Empty — No audio has been generated yet.
• Generating — Audio is currently being generated (⟳ spinner).
• Done — Audio has been successfully generated.
• Edited — The audio has been edited in the waveform editor.
• Stale — The text or voice has changed since the audio was generated.
• Error — Generation failed (hover for error details).

Re-generation

Segments become "stale" when their text or voice assignment changes after generation. Stale segments show an indicator and can be re-generated individually or via Generate All. Re-generating a segment will prompt for confirmation.

Timing Settings

Control the silence between audio segments in the exported audiobook:

• Between snippets — Default gap between consecutive segments.
• Glued snippets — Gap for segments connected with the glue toggle (typically zero or very short).
• Between chapters — Silence inserted between chapters in the full audiobook export.

Parallel Generation

By default, snippets are generated 2 at a time. You can adjust the parallel generation count in the Settings dialog (⚙). Higher values speed up generation but use more of your ElevenLabs API quota simultaneously.