Transcript timestamp inserter
Estimate and insert timestamps into a plain-text transcript based on word count and speaking rate.
Transcript timestamp inserter
Paste a transcript that has no timestamps and this tool estimates where each paragraph or sentence falls in the recording, based on word count and a speaking rate you choose. It's built for podcast editors, video creators and anyone turning a raw transcript into subtitles, show notes or YouTube chapters without re-listening to the whole recording to mark times by hand.
Set the speaking rate in words per minute to match your speaker — a slow narrator might sit around 110 wpm, a fast conversational podcast closer to 160. Choose whether a timestamp is inserted before every paragraph or every sentence, pick hh:mm:ss or mm:ss, and set a starting time if the transcript continues from an earlier point in the recording. A pause value adds a few seconds of estimated silence between paragraphs, which keeps longer stamps from drifting too far ahead of the real audio. Turn on YouTube chapters format to get a flat list of "mm:ss title" lines ready to paste into a video description — YouTube requires the first chapter to start at 0:00, so leave the start time at zero for that format.
This is an estimate, not a transcription tool: it has no access to the actual audio, so pauses, filler words and speaking-rate changes across the recording will shift the real times against the estimate. Align timestamps to whole seconds for a tidy hh:mm:ss line, or turn that off to see the fractional second the estimate lands on. The tool handles Windows (CRLF) and Unix (LF) line endings, and the live tally reports the word and paragraph or sentence count plus the estimated total duration.
Everything runs locally in your browser — your transcript is never uploaded anywhere. Copy the timestamped result, download it as a .txt file, or send the output back into the input to adjust the options and refine the estimate further.