I kept downloading tracks from early 2000s DJ CDs expecting high quality, only to find they sounded terrible. Turns out many of them are just low-bitrate MP3s ripped from YouTube and re-encoded into WAV or FLAC to fake quality. So I built avil — a macOS CLI tool in C++ that analyzes an audio file's frequency spectrum to detect whether it's a genuine high-quality recording or an upscaled fake. It also includes a real-time command-line spectrogram and volume visualizer. The detection works by running an FFT on the audio and checking for the characteristic high-frequency cutoff that compressed formats like MP3 leave behind, even after re-encoding. GitHub: https://github.com/giorgiogamba/avil Would love feedback on the detection approach and any ideas for improving accuracy.