Try Pro FREE for 2 weeksGet Started
Start for free

video-subtitle — Traditional AI Model

Burns karaoke subtitles onto each clip, highlighting each word as it is spoken, from per-word timings paired to the clip by index. CJK-capable rendering, composited with FFmpeg so the audio is untouched.

Try in Studio

Details

  • CategoryTraditional
  • Year-1
  • License

Inputs & Outputs

ClipsInputlist:video

Video clips to subtitle.

TranscriptsInputlist:transcript

Per-clip transcripts (matched to clips by index).

TranslatedInputlist:transcriptoptional

Optional second transcript (e.g. the same clips run through Speech Recognition with task=Translate to English) — stacked as a second subtitle line under the spoken one.

FontInputfile:fontoptional

Optional uploaded font (.ttf/.otf) to render the subtitles with; overrides Font Family when connected.

ClipsOutputlist:video

Subtitled clips

Parameters

positionstringdefault Auto (by shot type)options: Auto (by shot type), Bottom, Center, Top

Caption anchor. Auto places it per clip from the clip's shot_type (wide -> top, face/mouth/food -> bottom) so it stays off the subject; an explicit anchor is used for every clip and overrides the template's own.

Where the subtitle block sits. 'Auto (by shot type)' (default) decides PER CLIP from the shot_type the analysis stage attached: a wide shot puts a small subject low in frame so captions go to the top, while face/mouth punch-ins and food close-ups fill the upper-middle (and food already has a label track along the top), so captions stay at the bottom. A clip with no shot_type falls back to the template's position. Picking an explicit Top/Center/Bottom applies it to every clip AND overrides the template's built-in anchor - use that to keep captions clear of a busy region (e.g. source video with its own burned-in subtitles along the bottom).

templatestringdefault Customoptions: Custom, Minimal, Vibrant, Bold, Newsletter, Clean, +1 more

Look preset. A named template sets the font, accent color, position, box, outline (and animation) as one bundle (overriding the individual style options below). 'Bubbly (Bouncy Pet)' is a cute pet-shorts style (bold + thick outline + pink accent) that also pops each phrase in (bounce). 'Custom' uses the options below.

animationstringdefault Noneoptions: None, Bounce

Entrance animation for each phrase (Custom template only; presets set their own). 'Bounce' pops each phrase in with a cute spring settle (통통); 'None' shows it statically.

font_familystringdefault Noto Sans CJKoptions: Noto Sans CJK, Noto Sans CJK Bold, DejaVu Sans, +2 more

Subtitle font. The 'Noto ... CJK' faces render Korean/Japanese/Chinese; the Latin faces (DejaVu/Liberation/Ubuntu) are for English-only subtitles and show boxes on CJK text.

font_sizeintegerdefault 54range 20–120advanced

Subtitle font size (px).

max_chars_per_lineintegerdefault 16range 8–40advanced

Wrap each subtitle line to this many characters.

animation_strengthintegerdefault 100range 0–100advanced

Bounce intensity % (0 = no motion)

How strongly the bouncy pop-in springs, as a percentage of the tuned default. Only applies when the animation is Bounce (the Bubbly template turns it on automatically). The amplitude scales with frame height and is clamped so the caption can never leave the frame; lower this if the pop feels too energetic, set 0 for a static caption.

Tags

  • video
  • subtitle
  • caption
  • karaoke
  • transcript
  • shorts
  • list

Alternatives in Video Editing

  • Audio Edit

    Trim a sound, fit it to a video or a length, dip or mute parts, and fade it.

  • Audio Finish

    Normalize a video's loudness to a broadcast target (2-pass loudnorm + peak limiter).

  • Audio Mix

    Mix narration, music and sound effects into one track, with ducking and loudness.

  • Audio Replace / Mix

    Replace or mix a clip's audio with a new sound, keeping the video as is.

  • Chroma Compose

    Cut a green-screen person out of one video and place them over another.

  • Hook Teaser (Cold Open)

    Prepend a cold-open teaser of each clip's hook moment to the front.

  • Ken Burns (Still to Video)

    Animate a still image into a video with a moving crop window.

  • Narration Mux (Cold Open)

    Prepend a voiceover cold-open (frozen first frame + narration) to a clip.

All AI models