Self-Supervised Learning of Multi-Level Audio Representations for Music Segmentation.
Morgan BuissonBrian McFeeSlim EssidHélène C. CrayencourPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2024)
Keyphrases
- learning algorithm
- learning process
- supervised learning
- audio signals
- learning systems
- multiscale
- multimedia
- online learning
- music scores
- multiple representations
- region growing
- fully automatic
- knowledge acquisition
- machine learning
- image segmentation
- reinforcement learning
- edge detection
- level set
- object detection
- active learning
- prior knowledge
- learning tasks
- image processing
- bayesian networks