Introducing Song Creator Pro — create music with AI, locally on your device. Try it now →

Project Settings

Configure generation, segmentation, and output settings for your project.

Open Project Settings to configure generation, segmentation, and output options for the entire project. These settings act as defaults that can be overridden at the segment or selection level.

Generation

SettingDefaultDescription
LanguageAutoOutput language. Auto detects from the text.
Takes per generation1Number of takes to generate per segment. Extra takes appear in the segment's history so you can pick the best one.

Generation Parameters

Advanced generation parameters (speed, steps, guidance scale, etc.) can be configured at three levels:

  • Project level - Set defaults in Project Settings that apply to all segments
  • Segment level - Override parameters for individual segments
  • Selection level - Highlight text within a segment to adjust settings for that specific portion

The available parameters depend on the selected model. See the Voice Cloning docs for a breakdown of each model's settings.

Segmentation

SettingDefaultDescription
Max chars per segment500Maximum characters before a segment is auto-split. Set to 0 to disable auto-splitting. Paragraphs longer than this are split at sentence boundaries. Click Re-parse to apply changes to the current document.
Clean up imported text-Strips decorative symbols while normalizing ligatures and whitespace. Keeps punctuation, dashes, and quotes.

Output

SettingDefaultDescription
Paragraph gap (ms)300Silence between paragraphs in milliseconds.
Segment gap (ms)1000Silence between segments in milliseconds.
FormatMP3Export format (MP3, M4B, WAV, or FLAC).