The default: bottom centre, inside the safe area
Bottom-centre placement is the near-universal default for subtitles and captions. Viewers expect to find text there, it usually covers the least important part of the picture, and the eye can drop to it from faces in the middle of the frame and return quickly. Two-line subtitles typically grow upward from a fixed bottom line, so the lowest line stays in the same place and only the top line comes and goes.
Players apply this default automatically when a subtitle file contains no positioning. That is usually the right result, which is why most subtitle files never specify a position at all. Positioning becomes a decision only when the default would hide something the viewer needs.
Moving subtitles up to avoid burned-in text
The most common reason to move a subtitle is text already in the picture: name captions and lower thirds, on-screen titles, subtitles burned into archival clips, price tags in a product video, or the credits. If the subtitle sits on top of that text, both become unreadable.
Most professional guides handle this the same way:
- Move the subtitle to the top of the frame for exactly as long as the on-screen text is visible, then return it to the bottom.
- Do not move subtitles for text that carries no information the viewer needs, such as a small logo.
- Avoid shuttling subtitles up and down cue by cue; if on-screen text appears repeatedly in a short scene, keeping subtitles at the top for the whole scene is calmer.
- When the on-screen text is in another language and matters to the story, it may need its own translated subtitle, a topic covered in translating on-screen text in videos.
A 6-minute demo opens with a presenter whose name and job title appear as a lower third from 0:04 to 0:11, and a feature callout appears at the bottom of the frame from 2:30 to 2:38. The subtitler moves the three cues that overlap those spans to the top of the frame and leaves every other cue at the bottom. In the VTT version this is done with a line setting on those three cues; for the platform that only accepts plain SRT and ignores position tags, the presenter's cues are condensed so they end before the lower third appears, and the callout cue is checked by eye on the published video.
Title-safe areas and margins
Television once lost the edges of the picture to overscan, so broadcasters defined inner rectangles: an action-safe area for important picture content and a smaller title-safe area for text. Modern screens show the whole frame, but the idea survives because text placed hard against the edge is uncomfortable to read and can be cropped by some devices, projectors or platform layouts.
Safe areas are usually expressed as a percentage of the frame inset from each edge. The exact values differ between older standard-definition practice, current broadcast recommendations and individual platform specifications, so take the figure from the spec you are delivering to. Most editing applications can overlay safe-area guides, which makes checking burned-in subtitles quick.
For files rendered by a player, the player chooses the margins, and you mainly need to make sure your positioning does not push text outside the frame.
Vertical video and platform interface overlap
On vertical short-form platforms, the bottom of the frame is not empty. The platform overlays the creator name, the post's caption text, music credits, a progress bar and a column of buttons along one side. Subtitles placed at the very bottom, as they would be on a landscape video, often end up behind that interface.
Creators who burn subtitles into vertical video therefore usually place them higher, somewhere in the lower-middle part of the frame, and keep them away from the side where buttons sit. Several platforms publish safe-zone guidance or templates for this, but the interface changes with app updates and differs between platforms and devices, so check the current guidance and preview on a real phone before publishing. Uploaded caption files on those platforms are positioned by the app itself, which generally avoids its own controls.
Positioning in WebVTT: cue settings
WebVTT has positioning built into the format. Each cue's timing line can be followed by cue settings:
- line
- The vertical position, as a line number or a percentage from the top of the video area; for example, a small percentage places the cue near the top
- position
- The horizontal position of the cue box, as a percentage across the video area
- size
- The width of the cue box, as a percentage of the video width
- align
- Text alignment within the box: start, center, end, left or right
- vertical
- Vertical text for scripts written top to bottom, such as some Japanese text
- region
- Assigns the cue to a region defined in the file header, used for scrolling caption areas
Browsers implement these settings through their native caption rendering, and most support line, position and align. Regions and vertical text are supported less consistently. Custom web players may ignore settings entirely or apply them differently, and platforms that import VTT sometimes discard positioning. Test in the actual player your audience uses.
Burned-in versus player-rendered placement
Burned-in subtitles are placed once, at export, and stay exactly there on every device. That is an advantage for vertical social video and for screens where position must be guaranteed, but the viewer cannot move or hide them, and any mistake means re-exporting. The broader trade-off is explained in open vs closed captions.
Player-rendered subtitles are positioned by the player, which applies the file's positioning where it understands it and its own defaults otherwise. Many players and operating systems also let viewers change caption appearance, and some allow moving captions, which is an accessibility feature worth preserving. How text looks once it is positioned, from font and size to outline and background, is covered in subtitle readability.
Positioning mistakes and support risks
- Moving subtitles to the top for the whole video because of a logo or a single title card.
- Leaving subtitles over lower thirds, so viewers can read neither the name nor the line.
- Burning subtitles into vertical video at the very bottom, where the app's interface covers them.
- Relying on SRT position tags for a platform that strips or ignores them.
- Placing text near the frame edge, where it may be cropped on some displays.
- Forgetting that two-line subtitles grow upward and can collide with on-screen text higher in the frame.
Positioning mydubly subtitles
mydubly's SRT and VTT files contain no positioning: every cue is a single line of recognized speech with no cue settings or override tags, so players show it in their default bottom position. mydubly does not translate text that appears in the picture and does not burn subtitles into the video, so burned-in titles and lower thirds stay exactly where they were, and the subtitle file may need adjusting around them.
To position cues, open the file from the subtitle generator in a subtitle editor, find the cues that overlap on-screen text, and add a line setting in the VTT version or a top-placement tag in the SRT if your player supports it. A 6-minute video costs 6 credits (0.6¢) for transcripts and both subtitle formats, so it is cheap to generate a fresh file from the final cut before placing cues.
Next step: check every overlap
Play your video with subtitles on and pause wherever text appears in the picture. Note the time spans, move only the cues that collide, and preview the result on the platform where you will publish, on the device type most of your audience uses. For uploading the finished file, the YouTube subtitles use case walks through one common destination.
Frequently asked questions
Where should subtitles be positioned on screen?
At the bottom centre of the frame, inside a safe margin from the edges, unless they would cover important on-screen text. Viewers expect them there, and players use that position by default. Move them to the top only for as long as something important would otherwise be hidden.
How do I move subtitles to the top of the screen?
In a WebVTT file, add a line setting to the cue's timing line with a small percentage so it sits near the top. In SRT, many desktop players honour an override tag written as a backslash followed by an8 inside curly braces at the start of the cue text. Platforms vary, so test where you publish.
Do SRT files support subtitle positioning?
Not in the basic format. Some players understand unofficial override tags borrowed from the ASS format, and a rarely supported coordinate extension exists, but many platforms ignore both. Use VTT or burned-in subtitles when position must be guaranteed.
Where should captions go on vertical videos?
Higher than on landscape video, usually in the lower-middle part of the frame and away from the side where app buttons sit. The bottom of the frame is often covered by the creator name, caption text and controls. Interfaces change between apps and updates, so preview on a real phone.
What is the title-safe area for subtitles?
It is an inner rectangle of the frame within which text is guaranteed to be visible and comfortable to read. The idea comes from broadcast television, where screen edges were cropped. Exact margins differ between standards and platforms, so use the values in your delivery specification.
Can viewers change where subtitles appear?
Many players and operating systems let viewers adjust caption appearance, and some allow repositioning, but only for subtitles rendered by the player. Burned-in subtitles cannot be moved or hidden. That flexibility is one reason to deliver a separate subtitle file where possible.