CLIP BOT - CREDITS AND LICENSES
================================

Clip Bot itself is free software under the MIT license (see LICENSE.txt).

Clip Bot does NOT include the programs and AI models below in its download. They are fetched from their
official sources on first launch, or installed by pip, directly onto your PC. Each one keeps its own license.

PROGRAMS
--------
Python 3 ......................... PSF License             https://www.python.org
yt-dlp (YouTube downloader) ...... Unlicense (public domain) https://github.com/yt-dlp/yt-dlp
FFmpeg (BtbN "gpl" Windows build) . GPL v3                  https://ffmpeg.org  /  https://github.com/BtbN/FFmpeg-Builds
                                    (source code: https://github.com/FFmpeg/FFmpeg)
Deno (used by yt-dlp) ............ MIT                     https://deno.com
Ollama (local AI engine) ......... MIT                     https://ollama.com
faster-whisper ................... MIT                     https://github.com/SYSTRAN/faster-whisper
CTranslate2 ...................... MIT                     https://github.com/OpenNMT/CTranslate2
Silero VAD (inside faster-whisper)  MIT                     https://github.com/snakers4/silero-vad
huggingface_hub .................. Apache 2.0              https://github.com/huggingface/huggingface_hub
NumPy ............................ BSD-3-Clause            https://numpy.org
requests ......................... Apache 2.0              https://requests.readthedocs.io
CustomTkinter .................... MIT                     https://github.com/TomSchimansky/CustomTkinter
Pillow ........................... MIT-CMU (HPND)          https://python-pillow.org
RapidOCR (onnxruntime) ........... Apache 2.0              https://github.com/RapidAI/RapidOCR
  (uses PaddleOCR models) ......... Apache 2.0              https://github.com/PaddlePaddle/PaddleOCR
ONNX Runtime ..................... MIT                     https://onnxruntime.ai
OpenCV (opencv-python-headless) .. Apache 2.0              https://opencv.org
NVIDIA CUDA libraries (only on NVIDIA PCs, via pip: nvidia-cublas-cu12, nvidia-cudnn-cu12)
                                    NVIDIA proprietary license (free to use; downloaded from NVIDIA's own pip packages)

AI MODELS
---------
Speech: Systran/faster-whisper-small ............... MIT         (OpenAI Whisper, converted)
Speech: mobiuslabsgmbh/faster-whisper-large-v3-turbo  MIT         (OpenAI Whisper large-v3-turbo, converted)
Hebrew speech: ivrit-ai/whisper-large-v3-turbo-ct2 .. Apache 2.0  https://huggingface.co/ivrit-ai
Hebrew speech: ivrit-ai/whisper-large-v3-ct2 ........ Apache 2.0  https://huggingface.co/ivrit-ai
Local AI: Qwen3 (qwen3:8b / qwen3:4b / qwen3:1.7b) . Apache 2.0  https://huggingface.co/Qwen

FONTS
-----
Captions use Arial / Arial Black. They are copied from your own Windows fonts folder
(C:\Windows\Fonts) when you run Clip Bot. They are not shipped with Clip Bot.

ABOUT THE GPL (FFmpeg)
----------------------
FFmpeg's "gpl" build is GPL v3. Clip Bot only downloads it from its official build page and runs it as a separate
program. Clip Bot does not bundle or modify it, so Clip Bot itself can stay MIT. If you ever ship FFmpeg INSIDE
your own download, you must also offer its source code (the link above is enough).
