How to Match a Translated DOCX or TXT to SRT Timecodes

Published 2026-08-06 · FrameMath Guides

Translators often prefer Word documents, while editors need timed SRT files. The safest bridge between those workflows is to keep the original SRT as the timing master and map each translated DOCX paragraph back to the corresponding subtitle cue.

The FrameMath subtitle matching tool performs this mapping locally. It reads the SRT/VTT and DOCX/TXT in your browser, pairs text by cue order, lets you edit the result, and exports a timed subtitle file. The files are not uploaded.

Prepare the document before matching

Ask the translator to follow one rule: one non-empty paragraph per subtitle cue, in the original order.

The translated paragraph does not need the same number of words as the source. It only needs to correspond to the same cue. The translator may move punctuation or rewrite a phrase naturally, but merging and splitting paragraphs changes the alignment.

A clean TXT handoff might look like this:

城市在日出前苏醒。
第一班火车已经离站。
我们只有一次机会。

You can download a sample translated TXT and pair it with the sample English SRT.

Match DOCX or TXT to SRT

  1. Open the subtitle translator and bilingual SRT maker.
  2. Load the original SRT or VTT.
  3. Import the translated TXT/DOCX or paste the translated paragraphs.
  4. Select Auto-match translation.
  5. Compare the source cue count with the translated paragraph count.
  6. Repair mismatches in the cue editor.
  7. Export translation-only or bilingual SRT/VTT.

DOCX parsing happens locally. FrameMath extracts normal paragraph text; it does not upload the Word document or retain a copy.

What if the line counts do not match?

A difference of zero is the ideal automatic case. A difference of one does not mean “close enough” — it can shift every subsequent line onto the wrong cue.

Work from the first mismatch, not the last cue. Look for:

  • A blank translation where the source contains a sound effect
  • Two source cues translated as one paragraph
  • One source cue translated into two paragraphs
  • Headings, translator notes or comments included as paragraphs
  • A deleted or duplicated cue in the SRT

If two source cues were deliberately merged in translation, either split the translated sentence across the two cues or manually edit the timing in a subtitle editor. Order-based matching cannot infer creative restructuring reliably.

Translation-only versus bilingual export

Choose translation only for the final audience subtitle. Choose source + translation for review, linguistic quality control and client approval. A bilingual export makes omissions and meaning changes much easier to spot because both lines share the same timecode.

Before delivery, inspect cues containing:

  • Names, brands and places
  • Numbers, currencies and measurements
  • Dates and times
  • On-screen text
  • Jokes, idioms and wordplay
  • Very short display durations

These are the places where an apparently aligned document can still contain a serious translation or readability problem.

Fix the format before matching

The timing source should be a valid SRT or VTT. If your source is ASS, SSA, SBV or LRC, first use the subtitle format converter. If the timing itself is offset or drifting, repair it with the subtitle sync fixer before adding the translation. Fixing timing after bilingual export creates unnecessary duplicate work.

A reliable handoff checklist

  • Keep an untouched copy of the original timed subtitle.
  • Give the translator numbered or clearly separated cues.
  • Request one translated paragraph per source cue.
  • Remove document headings and translator notes before matching.
  • Resolve the first line-count mismatch before reviewing later cues.
  • Export a bilingual review version before the final translation-only file.
  • Watch the final subtitle over picture, not only in a text table.

This workflow separates responsibilities cleanly: the SRT owns timing, the translated document owns language, and the final review confirms they still correspond.

← All guides