🎬 Script Formatter

The Script Formatter breaks text after likely sentence punctuation, wraps unusually long lines at word boundaries, and can add extra paragraph spacing plus commas before selected conjunctions. Templates then add planning labels such as HOOK, INTRO, CONTENT, CTA, or END around the existing text. These labels help organize an editing draft; a TTS engine may read them aloud unless you remove them or generate each section separately.

The optional speaker-label setting does not detect who is speaking. With no template selected, it alternates non-empty lines between Speaker 1 and Speaker 2. That can be useful for laying out a two-person draft, but it is not dialogue recognition and it does not assign different voices automatically. Review every label, split each speaker into the intended voice workflow, and remove production notes before final synthesis.

The displayed duration is a simple word-count estimate based on roughly 150 words per minute and rounded up to whole minutes. Actual audio length changes with punctuation, language, selected voice, engine, speed setting, pauses, and the way numbers or symbols are spoken. Use the estimate to scope a script or identify an obviously overlong section, not as a frame-accurate promise for a video timeline, advertisement, or broadcast slot.

A reliable production check is to format one section, remove labels that should not be spoken, generate it with the final engine and speed, and compare the resulting waveform with the target timeline. Adjust the script based on measured audio rather than repeatedly forcing punctuation. Copy and Download preserve the formatted text; Send to TTS uses session storage to open it in the TTS workspace. The formatter does not save a project history, so keep your source document and a reviewed output copy.

For multilingual scripts, format each language block separately and verify punctuation conventions with a fluent reviewer. The generic sentence and conjunction rules were not designed to reconstruct every writing system. Keeping language blocks explicit also makes it easier to choose a compatible TTS voice and document when a phonemization request will occur.