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

Hearables with integrated microphones may offer communication benefits innoisy working environments, e.g. by transmitting the recorded own voice of theuser. Systems aiming at reconstructing the clean and full-bandwidth own voicefrom noisy microphone recordings are often based on supervised learning.Recording a sufficient amount of noise required for training such a system iscostly since noise transmission between outer and inner microphones variesindividually. Previously proposed methods either do not consider noise, onlyconsider noise at outer microphones or assume inner and outer microphone noiseto be independent during training, and it is not yet clear whetherindividualized noise can benefit the training of and own voice reconstructionsystem. In this paper, we investigate several noise data augmentationtechniques based on measured transfer functions to simulate multi-microphonenoise. Using augmented noise, we train a multi-channel own voice reconstructionsystem. Experiments using real noise are carried out to investigate thegeneralization capability. Results show that incorporating augmented noiseyields large benefits, in particular considering individualized noiseaugmentation leads to higher performance.

Multi-Microphone Noise Data Augmentation for DNN-based Own Voice Reconstruction for Hearables in Noisy Environments