Aegisub edits subtitle scripts as timed lines rather than changing the video itself. Each dialogue line has text, a start time, an end time, and a style reference. Aegisub can open the matching audio and video for timing and preview, but saving the subtitle script does not burn words into the picture. A separate player, renderer, or video-encoding step must apply the finished subtitles.
Dialogue lines
The subtitle grid holds dialogue, comments, style names, margins, actor fields, and effect data. Selecting a line connects that row to the edit box and the active audio range. This makes a feature-length script manageable as individual events instead of one continuous block of text.
ASS scripts preserve positioning, typography, outlines, shadows, and inline overrides. Simpler subtitle formats mainly preserve text and timing. Aegisub can open or export several formats, but the destination format decides which styling survives.
Audio timing
Aegisub displays a waveform or a frequency spectrum beside the active line. The start and end markers can be placed with separate mouse buttons or dragged after playback. Short playback controls cover the edges of the selection so the user can hear whether speech begins before the subtitle or continues after it.
Timing changes need a commit unless auto-commit is active. Moving to the previous or next line through the discard controls can throw away an uncommitted adjustment. Auto-next can select the following line after a normal commit, but an automatic commit does not trigger the same move. Those options change the rhythm of the timing workflow and can cause missed edits when their state is misunderstood.
Keyframe markers show likely scene boundaries when video data is available. Holding the relevant modifier snaps a line boundary to another line or a keyframe. The spectrum view replaces waveform height with frequency and uses color for signal strength, which can separate voices from steady background noise more clearly.
Audio caching
Compressed audio must be decoded to PCM before Aegisub can use it for precise timing. The default cache keeps that decoded stream in memory. A long film can therefore consume far more RAM than the compressed source file suggests.
A hard-drive cache reduces memory use but responds more slowly. Converting the track to an uncompressed WAV file avoids the same decoding cache step. Aegisub also works with mono audio internally. It downmixes multichannel sources, and material with more than two channels can produce a poor mix unless the user selects a suitable channel.
Script styles
An ASS style stores reusable font, color, outline, shadow, alignment, and margin rules. Assigning one style to many dialogue lines lets a later style edit update their shared appearance. Style Storage keeps reusable definitions outside the current script and copies selected styles into a project when needed.
The format has awkward boundaries. Some properties exist only as inline override tags, while others exist only in the style editor. A script cannot contain two styles with the same name, and ASS has no justified alignment. Opaque-box rendering can also behave incorrectly when font scaling differs from its normal value.
Export filters
Export differs from Save because it can transform the script before writing the new file. Filters can limit output to visible lines, generate karaoke effects, convert timestamps for another frame rate, remove redundant tags, clean script information, or replace missing styles. Aegisub applies selected filters in the order shown, so changing that order can change the output.
Saving to a format other than ASS usually strips many formatting tags. Frame-rate transformation can recalculate every timestamp, and variable-frame-rate conversion needs the correct video timing data. The Clean script info filter removes nonessential headers, which matters before distribution because an ASS project can store paths to the last opened audio or video. The exported subtitle file still needs a final playback check with the intended renderer.
