Cortex Plugin: Video & Audio Processor

Transcribe, translate, caption, clip video/audio files using FFmpeg and Whisper.

Installation

cortex plugin install github:CortexPrism/cortex-plugin-video-processor

Tools

video_transcribe

Transcribe video or audio to text/subtitles.

• file_path (string, required) — Path to video or audio file
• language (string, default: "en") — Language code
• output (enum) — Output format: text, srt, vtt, json
• model (enum) — Whisper model: tiny, base, small, medium, large

video_translate

Translate a transcript to another language.

• transcript (string, required) — Transcript text to translate
• target_language (string, required) — Target language ISO code
• source_language (string, default: "en") — Source language ISO code

video_generate_captions

Generate subtitle captions for a video file.

• file_path (string, required) — Path to video file
• language (string, default: "en") — Caption language
• format (enum) — Subtitle format: srt, vtt, ass

video_extract_clip

Extract a clip from a video.

• file_path (string, required) — Path to video file
• start_time (string, required) — Start timestamp (HH:MM:SS)
• duration (string, required) — Duration in seconds or HH:MM:SS
• output_path (string, required) — Output file path

video_extract_highlights

Extract a highlight reel from a video.

• file_path (string, required) — Path to video file
• transcript (string, required) — Transcript for highlight detection
• max_duration (number, default: 60) — Max highlight reel duration in seconds

Configuration

Key	Type	Description
ffmpegPath	text	Path to ffmpeg executable (default: "ffmpeg")
whisperModel	select	Default Whisper model: tiny, base, small, medium, large

Development

deno task test
deno fmt
deno lint

License

MIT