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 introduces a defense approach against end-to-end adversarialattacks developed for cutting-edge speech-to-text systems. The proposed defensealgorithm has four major steps. First, we represent speech signals with 2Dspectrograms using the short-time Fourier transform. Second, we iterativelyfind a safe vector using a spectrogram subspace projection operation. Thisoperation minimizes the chordal distance adjustment between spectrograms withan additional regularization term. Third, we synthesize a spectrogram with sucha safe vector using a novel GAN architecture trained with Sobolev integralprobability metric. To improve the model's performance in terms of stabilityand the total number of learned modes, we impose an additional constraint onthe generator network. Finally, we reconstruct the signal from the synthesizedspectrogram and the Griffin-Lim phase approximation technique. We evaluate theproposed defense approach against six strong white and black-box adversarialattacks benchmarked on DeepSpeech, Kaldi, and Lingvo models. Our experimentalresults show that our algorithm outperforms other state-of-the-art defensealgorithms both in terms of accuracy and signal quality.

Multi-Discriminator Sobolev Defense-GAN Against Adversarial Attacks for End-to-End Speech Systems