Caption timing is usually a cue clock: this line in, this line out. A word timestamp is finer. Each token gets a start, and often an end, so an accent can move with the mouth instead of with the phrase. Forced-alignment is the usual way to get that clock from a known transcript. ASR and speech-to-text may emit word times in the same pass, which is why a wrong word and a late word look like one failure until you read the file.
Karaoke captions and the Snap / Reveal families are the looks that spend this data. Ordinary subtitle families do not. SubRip has no standard per-word field; if the karaoke lives anywhere, it lives in the burn, in WebVTT with extra markup some players ignore, or in an editor's internal JSON. Do not expect an uploaded .srt to highlight tokens on YouTube.
The clock is only as honest as the transcript you fed it. Alignment will happily time your typo. Music and overlap smear the edges the same way they smear recognition. If one name is late, that is a local miss, not a reason to offset the whole track. A uniform slip after an intro trim is an offset of every timestamp, word-level included.
In practice
- Correct the transcript before you align. Word timestamps will treat your typo as spoken audio.
- Use word times for karaoke and word-highlight; use line times for ordinary subtitles.
- If the accent lags every word by the same amount, offset the track. If it drifts, look at frame rate, not at one token.
The mistake to avoid
Handing a line-timed SRT to a karaoke renderer and blaming the preset. The renderer cannot highlight words it was never told when they started.
Where you will run into it
- Add Captions to a Video — Speech in, styled subtitles out — no manual timing.
- Add Timed Text Overlays to a Video — Different words at different moments — you set the clock.
- AI Caption Generator — Drop a video, get timed captions. Free, on your device, no upload.
Related terms
Karaoke captions
Karaoke captions highlight each spoken word as it lands, using per-word timestamps instead of dropping a whole line on screen at once.
Caption timing
Caption timing is when each line appears and disappears relative to the speech — the clock, not the words and not the look.
Word-highlight captions
Word-highlight captions are burned-in subtitles that light up each word as it is spoken, using word-level timing and a caption-style accent colour.
Forced alignment
Forced alignment meaning: matching a known transcript to its audio to find exactly when each word was spoken.
Text-to-speech
Text-to-speech converts written text into spoken audio using a synthetic voice you choose before generating.
The all-in-one AI studio for creators. 60+ models for video, image, voice, music and lipsync, in your browser or on your phone.
Free on the web and on iPhone, one account.