What spotting means
Spotting, sometimes called cueing or timing, is the stage where a subtitler divides a video’s speech into subtitles and marks when each one should appear and disappear. The result is a list of empty or partly filled time slots. Each slot is a subtitle cue waiting for text, or already holding text.
The word “spotting” comes from marking spots in the video where something happens. The job is a mix of listening, watching and judging, and it affects readability as much as the wording does.
What the subtitler needs
To spot well, the subtitler needs a few things:
- The video itself, ideally the final edit, since later changes will move everything.
- The audio, clear enough to tell where speech starts and stops.
- A script or transcript, if one exists. It speeds up the work, though transcripts and scripts often differ from what is actually said.
- The style guidelines for the project, which set limits on duration, gaps, line length and reading speed.
- A subtitle editor, which shows a waveform or timeline and lets the person set start and end points while watching.
Having the right version of the video matters. If the video is re-edited after spotting, the timecodes will need adjusting, which is why project planning usually fixes the version before this stage begins.
The basic decisions
For each stretch of speech, the subtitler makes several judgments.
Where to split. Speech is cut at natural points: the end of a sentence, a pause, a change of speaker or a shift in thought. Cutting in the middle of a phrase makes subtitles harder to follow. Long sentences are broken at clause boundaries, so each cue reads as a coherent piece.
When to start. Typically a cue starts at, or slightly before, the first spoken word of its content. Starting too early gives away what is about to be said, while starting late leaves the viewer listening without reading.
When to end. A cue ends when the speech ends, plus a short extension if there is room, but not into the next cue or across a cut that makes it awkward. The end also depends on how much time the text needs.
How much text fits. The length of speech sets a time limit, and the time limit sets how many characters a viewer can read. If the speech is fast, the text has to be condensed. Some subtitlers think about this during spotting, so that the slots already reflect what can fit.
Speech, picture and sound together
Spotting is not only following the voice. A good spotter also watches the picture. Cuts in the video are natural places for cues to change, and a cue that crosses a cut can seem to flicker or to belong to the wrong scene. On-screen text, such as signs or titles, may need space or time. Music, laughter and other sounds may need a cue of their own if the subtitles are meant to include non-speech information.
Spotters also think about pace. A long run of very short cues can feel choppy. A run of very long ones can feel heavy. Variety is natural, but sudden extremes are worth reconsidering.
Templates and translation
In multilingual work, spotting is often done once in the source language, producing a timed template. Translators then write each language into the same slots. This saves effort and keeps all versions aligned, which makes it easier to manage many languages in one project. The trade-off is that different languages need different amounts of space. Some translations are longer than the source, so a template may be too tight in places. Text expansion therefore has to be handled, either by adjusting the template for some languages or by shortening the translation.
Translating from a spotted template also means the translator is often working on timed slots without seeing the whole sentence in its natural flow. Good templates keep sentences intact where possible and add notes for tricky spots.
Manual, assisted and automatic spotting
Spotting can be done by hand, which gives the most control, or with help from tools. Speech recognition can produce rough times for each word, and software can propose a first set of cues. This speeds up the early stages, but automatic timing usually follows the voice only, and it does not know about cuts, reading speed or natural phrase breaks. A person typically reviews and adjusts the result. Automated subtitling may reduce time on the first pass, though its results vary with audio quality and should always be checked.
Common mistakes in spotting
- Cues too short because the spotter followed each word exactly without considering reading time.
- Cues running across cuts when they could have been split.
- Splitting a phrase unnaturally, leaving a dangling word in one cue.
- No gaps, so successive cues blend together.
- Spotting before the edit is final, which can mean redoing the work.
Checking the result
Once the cues are placed, the subtitler plays the video with the timing alone, or with draft text, and watches whether the rhythm feels right. Does the text arrive in time with speech? Are any cues too brief or too long? Does every change land cleanly? Fixing these problems at this stage is much easier than after a full translation or review. When the timing is settled, writing and checking the text can proceed with some confidence that the slots will hold.