SRT to DOCX

Before and after on four cues

This is the SRT. Two speakers, four cues, one pause of 4.5 seconds.

1
00:00:01,000 --> 00:00:03,200
Maya: Thanks for joining. Let's start with the budget.

2
00:00:03,400 --> 00:00:05,900
Maya: I have the Q3 numbers up on screen.

3
00:00:07,000 --> 00:00:09,500
Dev: Can you zoom in on the travel line?

4
00:00:14,000 --> 00:00:16,000
Dev: Sorry, the travel line again.

We ran this exact file through this converter and read the text out of the .docx it wrote. Default settings (Show timestamps on, Merge same speaker on) give three blocks:

Above those blocks the document has its title and a grey summary line, here “3 blocks · 4 cues · SRT”. Two things happened. The cue numbers and --> lines are gone, and Maya’s two cues, 0.2 seconds apart, became one paragraph. Dev’s two cues stayed apart because the gap between them is longer than 2.5 seconds.

What each checkbox changes in that file

• Show timestamps off. The same three blocks, but the heading line is the speaker alone (“Maya”, “Dev”, “Dev”). A file with no speakers and no timestamps becomes plain paragraphs. The timecode page covers which readers want which.
• Merge same speaker off. Four blocks, one per cue: 00:01–00:03 Maya, 00:03–00:05 Maya, 00:07–00:09 Dev, 00:14–00:16 Dev. Use it when you need every cue boundary, for example to check line breaks against the video.
• Font, size, margins. Calibri 11 pt with 1-inch margins by default. Georgia, Times New Roman, Arial, and Cambria, 10–14 pt, and 0.75 or 1.25-inch margins are the other choices.
• Footer. Free exports carry a small “Made with SRT2Word · srt2word.com” source line. Pro exports use your own header and footer text instead.

If you upload a file instead of pasting, the download keeps the original name with .docx in place of .srt: episode-12.srt becomes episode-12.docx.

Look at the file first: three things we saw it do

The parser reads the format, not the meaning of the words. These are behaviours we saw when we tested it, not guesses:

• Italic tags and position codes are copied as text. A cue that says <i>Previously on the show...</i> or starts with {\an8} lands in the Word file with those characters visible. Delete them in a text editor with find and replace first.
• A bracketed sound cue can become a speaker. A cue that is only [Music] is read as a speaker named “Music”, because square brackets are one of the ways a speaker is written. Delete sound cues you do not want, or leave them if you do.
• Line breaks inside a cue depend on merging. With Merge same speaker off, a two-line cue stays two lines. When cues merge, line breaks become spaces.

A speaker only appears if the file names one, in the form Name:, [Name], or, in WebVTT, a <v Name> tag. The converter does not work out who is talking from the audio. For files with no names, see interview SRT to Word.

One file, or a folder

One file is free: paste it or drop it and press Download .docx. A season of episodes or a client folder is the Batch ZIP, part of Pro. Each file becomes its own .docx in one archive, with the same checkbox settings for every file. The unlock is a one-time payment, shown on the pricing page. Open one file here first and check the result before you run the folder.

If your file is not a plain SRT

• WebVTT from YouTube, a meeting, or a player: VTT to DOCX.
• Teams recap or Zoom cloud transcript: Teams, Zoom.
• Whisper output with loops and one-word cues: Whisper SRT to Word.
• Choosing between Word and a PDF to send: SRT to DOCX vs PDF.
• Moving the result into Google Docs: SRT to Google Docs.
• Step-by-step with screenshots of the options: how to convert SRT to Word.