The following notebooks evaluate the robustness of two audio fingerprinting algorithms: chromaprint and Qfp. The notebooks are self-explanatory and are required to be run in a Google Colab environment. The evaluations have been perfomed under three criteria:
- Robustness to superimposed noise.
- Robustness to pitch-shifting.
- Robustness to time-stretching.
The fingerprinting database has been created using the GTZAN Genre Recognition dataset.