Features

Video and images

Video concatenation — join multiple clips with optional xfade transitions
Image clips — Ken Burns effects (zoom-in, zoom-out, pan, smart, custom) with easing control
Image fitting — blur-fill, cover, and contain modes for aspect ratio mismatches
Color clips — flat colors and linear/radial gradients as first-class timeline elements with full transition support

Multi-source audio mixing — layer voiceovers, sound effects, and background tracks
Background music — loop support for tracks shorter than the video
Independent codec control — audio codec, bitrate, and sample rate are configurable at export

Text overlays — static, word-replace, word-sequential, and karaoke modes
Text animations — fade variants, pop, pop-bounce, typewriter, scale-in, pulse
Emoji support — opt-in emoji rendering via a custom font; stripped by default to prevent blank glyphs
Subtitle imports — SRT, VTT, ASS, SSA files with optional style overrides
Watermarks — text or image watermarks with preset or custom positioning and optional timed visibility
Effect clips — timed overlay effects with smooth fadeIn/fadeOut envelopes:

Effect	Description
`vignette`	Darkened edges
`filmGrain`	Noise overlay
`gaussianBlur`	Gaussian blur
`colorAdjust`	Brightness, contrast, saturation, gamma
`sepia`	Warm vintage tone
`blackAndWhite`	Desaturate to grayscale
`sharpen`	Sharpen detail
`chromaticAberration`	RGB channel split
`letterbox`	Cinematic black bars

probe() — full media metadata via ffprobe (duration, codec, fps, rotation, cover-art detection, and more)
snapshot() — capture a single frame to .jpg, .png, .webp, .bmp, or .tiff
extractKeyframes() — scene-change detection or fixed-interval frame sampling, returning Buffer[] or file paths
transcode() — hardened one-shot ingestion to web-safe H.264/MP4 (web-mp4) or AAC/M4A (web-audio)

audioTempo() — speed change that keeps pitch (real time-stretch, not resampling)
detectSilence() — silence intervals with tunable threshold and minimum length
spliceAudio() — rebuild audio from source ranges and inserted silence, with click-free micro-faded cuts
trimSilence() / capSilences() — trim edge silence, cap long interior pauses; cuts land inside silence, never on speech
normalizeLoudness() — two-pass EBU R128 normalization to a LUFS target

Platform presets — TikTok, YouTube, Instagram, Snapchat, and more
Auto-sequencing — omit position to place clips sequentially; use duration instead of end
Pre-validation — structured errors with machine-readable codes before any file I/O
Schema export — machine-readable clip spec for documentation, code generation, and AI pipelines
Progress callbacks — real-time percent, phase, fps, and speed during export
Live FFmpeg logging — onLog callback for raw FFmpeg stderr/stdout
Cancellation — AbortController support for stopping exports mid-render
Auto-batching — automatically splits oversized filter graphs to work within OS command-line limits
Custom temp directory — route intermediate files to fast storage (NVMe, ramdisk, etc.)
TypeScript — full type definitions included
Zero runtime dependencies — only requires FFmpeg on your system