Video to VTT

WEBVTT

00:00:02.100 --> 00:00:05.400

Here is the plan for this lesson.

lesson.vtt · ready for your player VTT

MP4, MOV, M4V, WEBM, MKV, or AVI, one video per job, up to 500 MB and about 60 minutes of audio - longer recordings are rejected up front with a clear message. We extract the audio track and transcribe the speech; if the file has no audio or no recognizable speech, the job stops and tells you.

Host the .vtt next to your video and reference it from a track element inside your HTML5 video tag, or upload it to your player or platform as the caption track. The file starts with the required WEBVTT header, so it works as-is.

Cue timing comes from silence-aligned speech chunks, which keeps captions in sync with the dialog for normal viewing. It is not word-level karaoke timing, so individual words are not highlighted as they are spoken.

No. Cues hold the spoken words shaped into standard two-line blocks of up to 42 characters per line, with no speaker names.

Subtitling is 5 credits per minute of audio. Because the length is not known until the file is processed, the upfront hold is estimated from file size at a typical video bitrate (about 8 MB per minute), so a typical minute works out to about 5 credits. The minimum is 5 credits.

Yes. Videos run on our own servers, are never used to train any model, and are auto-deleted 7 days after the job finishes.