Bass Single-Note Audio Dataset and Supporting Materials

This record contains the dataset and supporting materials used in a study on bass single-note recognition from audio recordings. The dataset consists of single-note electric bass recordings covering 52 note classes across four strings, ranging from the open strings to the 12th fret. The recordings were collected using a Precision Bass under consistent recording conditions. The record includes the raw audio dataset, extracted feature files, source code for feature extraction and model training, trained Support Vector Machine (SVM) models, and supporting documentation. The extracted audio features include Mel-Frequency Cepstral Coefficients (MFCC), Short-Time Fourier Transform (STFT), Constant-Q Transform (CQT), and pitch-related features. These materials are provided to support the reproducibility of the research and to facilitate further research in bass note recognition and Music Information Retrieval (MIR).