Speakers that stay readable
Detects [Name], Name:, and WebVTT <v Name> tags. Consecutive cues from the same speaker merge into paragraphs.
Convert Advanced SubStation Alpha (.ass) and SSA (.ssa) dialogue into an editable .docx. Speakers from the Name column, timestamps optional, karaoke and effects flattened to plain text. Nothing leaves your browser.
1 file free · Pro $4.99 one-time
Unlimited exports. Pay once.
Card, Apple Pay or Google Pay · No subscription
ASS is built for video players and typesetting: a [Script Info] header, styles, and timed
Dialogue: events with override tags for karaoke, position, and fonts. People ask for Word when
the script needs to leave the player — translation quotes, client review, or a deliverable specified as
.docx.
Opening an .ass in Word as plain text still leaves section headers, style rows, and
{\an8} markup. This converter reads Dialogue events only and writes real paragraphs
you can edit, comment on, and share.
.ass / .ssa / .txt file into the converter above.Name / Actor column as a speaker label.{\k}), italics/bold overrides, alignment
({\an}), {\pos}, fade, and font/style rows. You get timed
prose, not a restyled subtitle track.
{\p1} vector paths), Picture / Sound / Movie / Command lines.That honesty matches the job. Competitors that advertise “ASS to Word” still cannot put karaoke highlights into a useful Word document. We do not pretend otherwise.
Older SubStation Alpha scripts use ScriptType: v4.00 (no plus) and a slightly different Events
Format: line (Marked instead of Layer). The same parser accepts
.ssa on this page and on the homepage converter. SSA is covered here so we do not split one
searchable job across two thin URLs.
Word is not a timed caption format. Timestamps become plain reference text next to paragraphs — useful for humans, not for re-syncing a player by themselves.
.srt instead of Word: ASS to SRT, both directions.Built for real subtitle files and meeting transcripts — not a flattened plain-text dump.
Detects [Name], Name:, and WebVTT <v Name> tags. Consecutive cues from the same speaker merge into paragraphs.
Keep time codes for editors, or hide them for a clean reading script. Toggle updates the live preview instantly.
SubRip, WebVTT (including voice tags), ASS/SSA dialogue events, and common Zoom plain transcripts parse in the browser — no upload.
Clipboard HTML of the live preview, standalone HTML download, and print-to-PDF — plus .srt/.vtt/.ass drag-drop and local draft autosave. Nothing uploaded for conversion.
One transcript stays free, in this browser, with nothing uploaded to convert. Pro adds batch ZIP and header/footer only.
Dialogue events, SSA files, karaoke flattening, and what styling we intentionally drop.
ASS (Advanced SubStation Alpha, .ass) is the modern subtitle format used by Aegisub, fansub workflows, and many anime/video players. SSA (.ssa) is the older SubStation Alpha v4 script. Both store timed Dialogue events. This page owns the “ASS to Word” job; .ssa files use the same parser and the same converter — there is no separate SSA landing.
No — and that is intentional. Word is not a subtitle renderer. Override tags ({\b1}, {\an8}, {\pos}), karaoke ({\k}), drawings, and style/font metadata flatten to plain timed text. What you keep is dialogue, optional timestamps, the Events Name/Actor speaker, and readable paragraphs.
Dialogue (and text) events only. Comment, Picture, Sound, Movie, Command, and vector drawings are skipped. Consecutive cues from the same speaker merge when Merge same speaker is on.
Yes. Toggle Show timestamps off for a reading script. Keep them on when translators or reviewers need to jump back to the video.
No. Parsing and .docx generation run in your browser. See the Privacy Policy.