@symbolk

video-to-text transcribes video and audio files to plain text and srt subtitles using local whisper, supports 50+ languages with auto-detection, requires ffmpeg and openai-whisper, no api key needed for transcription alone.

video-resize reformats video to any aspect ratio via local ffmpeg center-crop (no api key), with optional escalation to ai-powered subject-aware cropping for platform-specific output.

video-clip extracts time-range segments from video files using local ffmpeg (no api required), with optional escalation to sparki ai edit for intelligent highlight detection and short-form content generation.