GISE-51 is a datasetspanning 51 isolated sound events belonging to a broad domain of event types . We also release a dataset of 5-second soundscapes with hard-labelled event boundaries synthesized from Gise-51 isolated sound . We conduct baseline sound event recognition (SER) experiments on the GISE51-Mixtures dataset, benchmarking prominent convolutional neural networks,and models trained with the dataset demonstrate strong transfer learning performance on existing audio recognition benchmarks . We hope to address some of the shortcomings of recent soundevent datasets, providing an open, reproducible benchmark for future research . For more information, visit www.gise.com/gise51-51 .

Author(s) : Sarthak Yadav, Mary Ellen Foster

Links : PDF - Abstract

Code :
Coursera

Keywords : gise - dataset - sound - isolated - event -

Leave a Reply

Your email address will not be published. Required fields are marked *