Transcript with Timestamps

What is a timestamped transcript?

A timestamped transcript is a readable text file, not a subtitle file: each line is a sentence or phrase prefixed with when it was said, e.g. [1:02:15] Let's move to the results.

What the file looks like

[0:00] Today we continue from the last session on supply and demand.
[0:18] Price is the signal, not the cause, of the change in quantity.
[0:41] Write down the three assumptions before the example.

Where it's used

• quoting interviews and meetings accurately
• drafting YouTube chapters
• finding moments in long lectures
• research coding and notes

Tips

• Timestamps come from the speech segments, so they line up with the subtitles exactly.
• For YouTube chapters, pick the lines where a new topic starts and paste them into the description as 0:00 Intro, 4:12 Topic, and so on.
• Choose a target language to get the timestamped transcript translated as well.

How the file is laid out

The timestamped transcript is a .txt file with one speech segment per line. Each line opens with the segment's start time in square brackets, then the words: minutes and seconds early on, and hours added once the recording passes the hour mark. There are no end times, cue numbers or blank lines between entries, which is what separates it from an SRT. One line looks like this:

Segments follow natural pauses, so line length varies: a fast talker produces long lines, a deliberate speaker short ones. In transcript mode the result screen offers it as Download timestamped transcript. Pick a target language and the lines are translated with their times kept, which makes it easy to go from a translated quote back to the original audio. The times match the SRT and VTT from the same job.

Quoting and citing a recording

Quotation is where timestamps earn their keep. Find the passage in the text, jump to that time in the recording, and listen before you quote; the transcript is a draft, and names, numbers and negations are the words most worth confirming by ear. Academic styles cite time-based media with a timestamp: APA, for example, asks for the time a direct quotation begins, in place of a page number, while MLA gives the time range of the quoted passage in the in-text citation. Because the file only records where each segment starts, take the end of a range from the next line's time or by ear. Keep the bracketed time with every quote in your notes and the citation writes itself. Quoting videos in research gives APA and MLA patterns.

Converting times into chapters, links and columns

YouTube chapters

the first chapter must start at 0:00, and YouTube expects at least three chapters, each 10 seconds or longer

Link to a moment

add t=754 to a YouTube link to start playback at 12:34

Times to seconds

hours × 3600 + minutes × 60 + seconds

Strip every timestamp

find-and-replace the regular expression ^\[[0-9:]+\]\s* with nothing, in any editor that supports regex

Spreadsheet columns

replace ^\[([0-9:]+)\]\s* with the captured time followed by a tab, then paste into a sheet

A column of times also makes clip hunting fast: scan the text for a strong line, note its time and cut from just before it (finding clips with transcripts).

Recordings longer than two hours

Files are limited to 2 hours, so a three-hour conference day or hearing goes in as two or more parts, and every part restarts at [0:00]. Before merging, add each part's start time to its lines (once times are in a spreadsheet column that's a single formula), or at least write "Part 2 starts at 1:30:00" at the top. One spreadsheet trap: Excel and Google Sheets read a minutes-only value such as 47:31 as 47 hours and 31 minutes, so prefix 0: to those times before doing any arithmetic. Cut at a pause rather than mid-sentence so no segment is split across files; splitting a video file shows how to cut without re-encoding.

Making the plain text more useful

Speakers aren't identified, so add initials after the time as you review, as in [12:04] JS:, which keeps the time reference intact. Notes of what's shown on screen at a given time turn it into a descriptive transcript for people who can't watch. When you hand the text to someone who only wants to read, strip the times with the expression above and tidy the paragraphs (formatting a transcript). Students revising from recorded classes will find the lecture transcription page useful, and if you need start and end times for a player, use the SRT generator.

Using the tool on this page

• The spoken language is detected automatically; you select the output language — the spoken language itself for an untranslated transcript, or another language for a translation. Files can be up to 2 hours long.
• Only the audio is sent for processing — the video stays on your device — and audio and text are deleted within 30 minutes of delivery (how files are protected).
• You can download a timestamped transcript, subtitles (SRT and VTT) and plain text. Every output, step and limit is explained on the subtitle generator page.