SALSA: Spatial Cue-Augmented Log-Spectrogram Features for Polyphonic Sound Event Localization and Detection.
Thi Ngoc Tho NguyenKarn N. WatcharasupatNgoc Khanh NguyenDouglas L. JonesWoon-Seng GanPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2022)
Keyphrases
- false positives
- spatial information
- event detection
- spatial relationships
- musical instrument
- spatio temporal
- feature set
- space time
- spatial features
- feature extraction
- feature space
- object recognition
- detection method
- visual features
- co occurrence
- spatial data
- image features
- low level
- spatial relations
- feature vectors
- extracted features
- spatial distribution