English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Audio classification is the task of identifying the sound categories that areassociated with a given audio signal. This paper presents an investigation onlarge-scale audio classification based on the recently released AudioSetdatabase. AudioSet comprises 2 millions of audio samples from YouTube, whichare human-annotated with 527 sound category labels. Audio classificationexperiments with the balanced training set and the evaluation set of AudioSetare carried out by applying different types of neural network models. Theclassification performance and the model complexity of these models arecompared and analyzed. While the CNN models show better performance than MLPand RNN, its model complexity is relatively high and undesirable for practicaluse. We propose two different strategies that aim at constructinglow-dimensional embedding feature extractors and hence reducing the number ofmodel parameters. It is shown that the simplified CNN model has only 1/22 modelparameters of the original model, with only a slight degradation ofperformance.

Reducing Model Complexity for DNN Based Large-Scale Audio Classification