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

Audio is an important medium in people's daily life, hidden information canbe embedded into audio for covert communication. Current audio informationhiding techniques can be roughly classed into time domain-based and transformdomain-based techniques. Time domain-based techniques have large hidingcapacity but low imperceptibility. Transform domain-based techniques havebetter imperceptibility, but the hiding capacity is poor. This paper proposes anew audio information hiding technique which shows high hiding capacity andgood imperceptibility. The proposed audio information hiding method takes theoriginal audio signal as input and obtains the audio signal embedded withhidden information (called stego audio) through the training of our privateautomatic speech recognition (ASR) model. Without knowing the internalparameters and structure of the private model, the hidden information can beextracted by the private model but cannot be extracted by public models. We usefour other ASR models to extract the hidden information on the stego audios toevaluate the security of the private model. The experimental results show thatthe proposed audio information hiding technique has a high hiding capacity of48 cps with good imperceptibility and high security. In addition, our proposedadversarial audio can be used to activate an intrinsic backdoor of DNN-basedASR models, which brings a serious threat to intelligent speakers.

Adversarial Audio: A New Information Hiding Method and Backdoor for DNN-based Speech Recognition Models