University of Twente Student Theses
Scrutinizing Sounds: A Study on Reconstruction Based Anomaly Detection in Vinyl Records’ Audio Data
Jonkers, J.W. (2023) Scrutinizing Sounds: A Study on Reconstruction Based Anomaly Detection in Vinyl Records’ Audio Data.
|
PDF
1MB |
| Abstract: | Anomalous sound detection is an important task in many real-world applications such as surveillance, quality control, and healthcare. For this research, we propose a novel application for state-of-the-art anomaly detection techniques. The proposed application is the detection of anomalous sounds within the recordings of vinyl records, which poses significant challenges due to the context-sensitive nature of the music stored on these records. This application poses further challenges, as due to the laborious process of labelling no labels were available for the used datasets. Considering these limitations use was made of a reconstruction-based detector featuring a denoising autoencoder (DAE) which uses a Long Short-Term Memory (LSTM) recurrent neural network (RNN). The study will investigate features which may serve to detect such anomalous sounds and the effectiveness of using a reconstruction-based anomaly detector for the stated problem. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science BSc (56964) |
| Link to this item: | https://purl.utwente.nl/essays/96402 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page