what is pitchcap fr?

PitchCap is a work in progress screen recording and video editing app for Mac, combining a Rust backend with a React and TypeScript interface. You start a recording with a keyboard shortcut or a menu bar icon, drag to select the part of the screen to capture, and choose whether to also record your camera, microphone, and system audio, with system audio needing an extra loopback tool installed separately. While recording, a pulsing red outline marks the capture area along with a small readout showing elapsed time and a stop button. When you stop recording, the editor opens automatically. The microphone track is transcribed into text using a speech recognition model, and the resulting words appear along the timeline, letting you select a stretch of spoken words and cut that section out of the video directly, a feature the project calls cropping by text. The camera recording appears as its own movable, resizable layer on top of the screen recording, and can be deleted and brought back later. Standard editing actions are supported too, including splitting a clip, trimming its edges, deleting a section, changing playback speed per segment, cropping the frame, and undoing changes. Exporting combines all these edits, the cuts, the camera overlay, and the mixed audio, into a single video file. Running it during development requires installing ffmpeg and a speech recognition tool called whisper.cpp, plus downloading a roughly 148 megabyte speech recognition model file the first time. Recording also requires granting macOS permission for screen recording, microphone, and camera access. Finished recordings are saved to a folder in the user's Movies directory. The project is licensed under the Apache License 2.0.