✏️ Punctuation Enhancer

The Punctuation Enhancer uses visible, repeatable text rules. It can normalize curly quotation marks, collapse repeated punctuation, capitalize letters after likely sentence boundaries, and add commas around selected transition words. The optional TTS pause rule also adds punctuation before words such as “but,” “yet,” and “so,” and converts paragraph gaps into section markers. It does not call a language model or infer the author's intended grammar from context.

Rule-based changes are fast and private, but ambiguity remains. A capitalized word after a space may be a name rather than a new sentence; a transition word may not need a comma; dialogue, poetry, code, legal clauses, and abbreviations can break ordinary sentence assumptions. The tool keeps the original text in the left box so you can compare every change. Treat the result as an editable draft, especially when punctuation affects legal meaning or quoted speech.

Automatic pauses are most useful for plain narration that already has clear sentences but feels rushed when synthesized. Try a short representative paragraph first, listen with the selected voice, and add only the pacing rules that improve it. A comma can create a brief break, while the inserted section marker creates a stronger boundary in workflows that recognize it. Voice, engine, speed, and chunking still influence the final rhythm, so punctuation alone cannot guarantee natural delivery.

Avoid applying every option blindly to subtitles, tables, source code, stage directions, multilingual text, or copy that must preserve exact punctuation. For those inputs, normalize quotes or repeated marks separately and leave sentence insertion disabled. Copy the reviewed result or use Send to TTS, which transfers text through browser session storage to the TTS workspace. If you later choose non-English Kokoro, that TTS step has the phonemization data path described in the Privacy Policy; this punctuation page itself processes the entered text locally.