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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This paper presents a way of doing large scale audio understanding withouttraditional state of the art neural architectures. Ever since the introductionof deep learning for understanding audio signals in the past decade,convolutional architectures have been able to achieve state of the art resultssurpassing traditional hand-crafted features. In the recent past, there hasbeen a similar shift away from traditional convolutional and recurrent neuralnetworks towards purely end-to-end Transformer architectures. We, in this work,explore an approach, based on Bag-of-Words model. Our approach does not haveany convolutions, recurrence, attention, transformers or other approaches suchas BERT. We utilize micro and macro level clustered vanilla embeddings, and usea MLP head for classification. We only use feed-forward encoder-decoder modelsto get the bottlenecks of spectral envelops, spectral patches and slices aswell as multi-resolution spectra. A classification head (a feed-forward layer),similar to the approach in SimCLR is trained on a learned representation. Usingsimple codes learned on latent representations, we show how we surpasstraditional convolutional neural network architectures, and come strikinglyclose to outperforming powerful Transformer architectures. This work hopefullywould pave way for exciting advancements in the field of representationlearning without massive, end-to-end neural architectures.

Attention is All You Need? Good Embeddings with Statistics are  enough:Large Scale Audio Understanding without Transformers/ Convolutions/  BERTs/ Mixers/ Attention/ RNNs or ....