SEPARATE TRACKS. ONE CONVERSATION.
Every voice,
in order.
Turn per-speaker SRT and VTT files into one readable transcript. Name each voice, align the clocks, and keep the moments people talk over each other.
01 / SOURCES
Bring your voices together
One file per speaker. You can mix formats.
UTF-8 · 10 MiB per file · 25 MiB total · up to 100 filesEverything stays in this tab. Refreshing clears your files.
Offsets are in seconds. Positive = later, negative = earlier.
02 / MERGED OUTPUT
The whole conversation
A place for every speaker
Add your files or try the two-speaker example.
Nothing gets uploaded, even in the demo.
SRT keeps original cue intervals, including overlaps. Some players show only one overlapping cue. TXT and Markdown are best for reading; JSON keeps source text, identifiers, metadata and timings.
Export timed captions
Start with one SRT or VTT for each speaker, using the same recording timeline.
Label & align
Filenames become speaker names. Edit labels and add an offset if a track starts late.
Take the conversation
Download a chronological transcript with all cues intact. Simultaneous speech stays simultaneous.