Licenças de código aberto

O Vidmini é feito com os seguintes softwares de código aberto. Obrigado aos seus autores. Os textos completos das licenças estão disponíveis em cada projeto.

• MediabunnyMPL-2.0

  Media demuxing, muxing and conversion

• @mediabunny/mp3-encoderMPL-2.0 (LAME: LGPL-2.0)

  MP3 encoding (WebAssembly)

• @mediabunny/aac-encoderMPL-2.0 (FFmpeg AAC encoder: LGPL-2.1)

  AAC encoding where the browser has none

• ffmpeg.wasm core 0.12.10 (FFmpeg, x264, x265, LAME, libvpx…)GPL-2.0-or-later

  Fallback converter for formats browsers cannot read (loaded only when needed). Corresponding source code: see the linked repository and FFmpeg source at https://ffmpeg.org/download.html; on request we provide the exact source of the files we distribute.

• Transformers.js (Hugging Face)Apache-2.0

  On-device speech recognition pipeline

• ONNX Runtime WebMIT

  Runs AI models in WebAssembly

• Whisper tiny / base (OpenAI, ONNX by onnx-community)MIT

  Speech-recognition models for automatic subtitles

• Ultra-Light-Fast-Generic-Face-Detector-1MB (RFB-320, ONNX Model Zoo)MIT

  Face detection for automatic face blurring

• gifencMIT

  GIF encoding

• JSZipMIT / GPL-3.0 (dual)

  ZIP archives of extracted frames

• AngularMIT

  Web framework

• Tailwind CSSMIT

  Styling

• Inter, Space Grotesk (Fontsource)OFL-1.1

  Fonts