START WITH THE TASK
Create video clips
Clipping has several stages: choose a moment, cut the media, transcribe speech and prepare the result. These local building blocks handle individual stages. None of this collection is presented as a complete automatic highlight-selection pipeline.
Where to start
For manual lossless trimming, begin with LosslessCut. Add Whisper when you need a draft speech transcript. Evaluate ffmpeg.wasm when you are developing media processing inside a browser application.
Compare the building blocks
Documentation-based guidance, not hands-on tests. Source-check dates belong to each fact; publishing this page does not refresh them.
| Repository and role | Setup | Execution and costs | Limitations | License and source |
|---|---|---|---|---|
| mifi/lossless-cut Trim video without the long render. | Download an official release for the OS, extract/run it (macOS DMG to Applications), or use an official app store. Settings and caches persist in the app-data directory; manual downloads are not fully portable.Checked 2026-11-20 | Execution requirements need review GPL-2.0-only app with free official GitHub downloads. Mac/Microsoft Store builds are paid to support development and provide automatic updates; underlying FFmpeg and bundled assets retain separate notices. No paid service is required for local editing.Checked 2026-11-20 | Lossless cut starts follow keyframes and may begin early; frame-exact cuts can require re-encoding. Smart Cut is experimental and not reliable for every file. Merging requires compatible codec parameters; app-store platform restrictions and preview codec limits apply.Checked 2026-11-20 | GPL-2.0 |
| openai/whisper The speech model behind many transcription tools. | Install openai-whisper with Python/PyTorch and the system FFmpeg command-line tool. The README describes expected compatibility with Python 3.8–3.11 and recent PyTorch versions; Rust may be needed where a tokenizer wheel is unavailable. Model weights are loaded separately by the library.Checked 2026-11-20 | Execution requirements need review The repository explicitly releases both Whisper code and model weights under MIT. Local inference still uses compute and storage; this license does not define pricing for a separate hosted transcription service.Checked 2026-11-20 | The model card warns of hallucinated or repetitive text and uneven accuracy across languages, accents and dialects. It is not a real-time transcription system out of the box. Turbo is not trained for translation and returns the original language even when translation is requested.Checked 2026-11-20 | MIT |
| ffmpegwasm/ffmpeg.wasm FFmpeg for browser, powered by WebAssembly | Install @ffmpeg/ffmpeg and @ffmpeg/util, host the worker-capable wrapper, then load compatible core JavaScript/Wasm assets. Multi-thread use requires SharedArrayBuffer security prerequisites. The usage guide distinguishes ESM assets for Vite from UMD examples.Checked 2026-11-20 | Execution requirements need review The JavaScript wrapper is MIT-licensed, but the compiled core follows FFmpeg and each included external library's licenses. Processing uses client compute; asset delivery/hosting remains separate, and the wrapper's license does not replace core/codec obligations.Checked 2026-11-20 | The FAQ describes WebAssembly performance as slower than native FFmpeg and does not claim RTSP support. Multi-thread mode has additional resource/security requirements. Progress reporting is experimental and may fail for operations such as concatenation or image conversion.Checked 2026-11-20 | MIT |
Before you commit
Try footage you have permission to use. Inspect cut boundaries, audio, transcript accuracy and exported playback. Browser processing has memory limits; a transcript still needs caption timing and review.