What these tools are for
Video files are large and their formats are messy. A phone recording might be too big for Discord or WhatsApp, a screen capture might be in WebM when the recipient needs MP4, and a two-hour recording might only need its first ten minutes. Audio has the same problems on a smaller scale: a voice memo in M4A, a song that should become a ringtone, a lecture that needs a transcript.
The tools are grouped into video, audio, speech and recording. Each one shows the result before you download it, reports progress with a time estimate and can be cancelled at any point.
How the processing works
Modern browsers include hardware video encoders and decoders through an interface called WebCodecs. Where your browser supports the format, the tools use it directly, which is often faster than real time. When a format is not supported natively, they fall back to FFmpeg compiled to WebAssembly, loaded only at the moment you start a job. Either way the video is read from your disk into the browser tab and written back as a download; it is never sent to a server.
Speech recognition for subtitles and transcripts uses an open Whisper model that runs in the browser, so interviews, meetings and voice notes stay private.
Limits
Processing speed depends on your device: a recent laptop compresses a short 1080p clip in seconds, while an older phone takes longer. The free plan handles files up to 500 MB and up to ten minutes of media per job, which suits most clips and recordings. Pro raises the duration and size limits for long recordings and larger batches.
Which tool to use
To send a clip that is over a platform's limit, use the compressor with a target size such as 10 MB for Discord or 16 MB for WhatsApp. If only part of a video matters, trim it first, because a shorter file compresses better. Converting is the right choice when a device refuses to play a format, and extracting audio turns a recorded talk into a lightweight file for listening.