Abstract
dc:description.abstractThis project explores the usage of a late fusion deep learning architecture to predict the geographic origin of music. Mel-Frequency Cepstral Coefficients (MFCCs) and the language of the music sample are used as features. MFCCs were extracted from audio files to capture sound features. The language was identified using OpenAI’s Whisper model to provide additional context. A late fusion neural network architecture combining Long Short-Term Memory (LSTM) layers for sequential MFCC input and dense layers for non-sequential language features were employed to support both classification and regression tasks. The classification model achieved an accuracy of 33.03% across 56 countries or territories, substantially outperforming the random baseline (1.79%). The regression model produced a Mean Great-Circle Error (MGCE) of 2,754.83 km. While regression offers more flexibility for geographic estimation, classification demonstrated a more promising performance with the current dataset. This work highlights the potential of multimodal learning in music-origin prediction.
Degree
thesis:*- Name thesis:degree_name
- Master of Science in Computer Science
- Level thesis:degree_level
- Project
- Discipline thesis:degree_discipline
- School of Computer Science and Engineering
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Ladanyi, Fruzsina
- Contributors dc:contributor
-
- Haiyan Qiao
Subjects
dc:subject × 12Identifiers
dc:identifier.*- Repository record dc:identifier
- https://scholarworks.lib.csusb.edu/etd/2303