Skip to content
SkillModel-led259 tokens in use

remotion-captions

Transcribe audio with local Whisper, import SRT files and display word-timed captions in a Remotion video.

7,648 stars on the repositoryremotion-dev/skills · skills/remotion-captions
npx skills add remotion-dev/skills
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing remotion-captions costs

14tokens idle
259tokens in use
3,555tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Fair
  • Frontmatter8/15
  • Structure5/25
  • Examples9/15
  • Output5/15
  • Size13/15
  • Resources11/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity88
  • Maintenance48
  • Community52
  • Maturity32
  • Momentum40

Captions guidance for Remotion projects. Every caption is JSON in the Caption type from @remotion/captions, with text, start and end times in milliseconds, a timestamp, and a confidence value.

The skill covers three jobs: transcribing a video or audio file into captions with Whisper running on the local GPU, importing an existing .srt file, and showing the captions with the Basic Captions element so they stay editable in Remotion Studio.

What You Get

  • A single JSON caption format (text, startMs, endMs, timestampMs, confidence, pageBreakAfter) used for every step
  • A Node.js transcription script that decodes audio to 16kHz mono and runs Whisper through @remotion/whisper-webgpu
  • A browser transcription variant using resampleTo16Khz for a File object
  • An SRT import component built on parseSrt, with a delayRender-guarded fetch
  • Rules for placing the Basic Captions element inside the same Sequence as its clip so trimming and speed changes affect both

Common Use Cases

  • Adding subtitles to a talking-head video built in Remotion
  • Converting a client-supplied .srt file into captions that animate with the video
  • Transcribing each source clip of an edit and keeping timestamps relative to the source file
  • Building a captioned short-form video where captions stay editable in Studio

What Makes It Unique

It is specific to Remotion’s own Caption type and the @remotion/captions package. It transcribes with a Whisper build that runs on the local GPU, so no audio leaves the machine, and it contains layout rules that keep captions attached to their clip when the clip is trimmed, reordered or retimed. It also asks for the captions array to be written inline so the Studio caption editor can write edits back to the source.

Under The Hood

The skill is instructions with code samples; it ships no scripts of its own. The agent is told to install packages with npx remotion add (or the bun, yarn or pnpm equivalent), write and run a Node.js script that reads a media file from the project’s public folder, downloads a Whisper model, and writes a captions JSON file. It tells the agent to fetch the Basic Captions source from remotion.dev and save it into the project, and it can fetch a remote or local .srt file. It also tells the agent to open Remotion Studio when finished. A compatible GPU is required for transcription.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers