Clarity-2021 Challenges: Machine Learning Challenges for Advancing Hearing Aid Processing

Graetzer, Simone; Barker, Jon; Cox, Trevor J.; Akeroyd, Michael A.; Culling, John F.; Naylor, Graham; Porter, Eszter; Muñoz, Rhoddy Viveros

doi:10.21437/interspeech.2021-1574

Cited by 48 publications

(23 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We expect that the fused models will emphasize latent target feature representations in the fused layers by canceling unwanted noisy elements contained in the input audio signal, causing also a decrease in layer activation variance. This work extends a previous study 1 presented at the 2021 Clarity speech enhancement challenge [36] by formalizing the concept and by analyzing the effect of input data correlation, latent activation variance, and encoding methods.…”

Section: Introductionmentioning

confidence: 59%

Deep Latent Fusion Layers for Binaural Speech Enhancement

Gajęcki¹,

Nogueira²

2023

Preprint

View full text Add to dashboard Cite

<p>In this work, we address the problem of speech enhancement in the context of binaural hearing. We propose deep learning models which are connected by ``fusion layers" that perform Hadamard products between specific generated latent representations. Fusion layers are inspired by multi-task learning approaches that combine and/or share weights between models that tackle related tasks. We first present a general fusion model and show that this approach is able to fit synthetic data better than independent linear models, equalize activation variance between learning modules, and exploit input data redundancy to improve the training error. We then apply the concept of fusion layers to enhance speech in binaural listening conditions. Our results show that the proposed approach improves speech enhancement performance on unseen data with respect to the independent models. However, we observe a trade-off between speech enhancement performance and predicted speech intelligibility based on a short-time objective binaural speech intelligibility index, potentially due to distortions introduced by fully fused models.</p> <p>Results also suggest that fusion layers should share parameterized latent representations in order to properly exploit the information contained in each listening side. In general, this work shows that sharing information between speech enhancement modules may be promising to improve binaural speech enhancement while keeping the number of trainable parameters constant and improving generalization. </p>

show abstract

Section: Introductionmentioning

confidence: 59%

Deep Latent Fusion Layers for Binaural Speech Enhancement

Gajęcki¹,

Nogueira²

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…In the clarity enhancement challenge [ 116 ], the task is to improve speech in the context of hearing aids. In contrast to the DNS separation challenge, the main objective was to improve intelligibility instead of quality.…”

Section: Techniques Used By the Winning Systems In Dns And Clarity Ch...mentioning

confidence: 99%

A Survey on Low-Latency DNN-Based Speech Enhancement

Drgas

2023

Sensors

View full text Add to dashboard Cite

This paper presents recent advances in low-latency, single-channel, deep neural network-based speech enhancement systems. The sources of latency and their acceptable values in different applications are described. This is followed by an analysis of the constraints imposed on neural network architectures. Specifically, the causal units used in deep neural networks are presented and discussed in the context of their properties, such as the number of parameters, the receptive field, and computational complexity. This is followed by a discussion of techniques used to reduce the computational complexity and memory requirements of the neural networks used in this task. Finally, the techniques used by the winners of the latest speech enhancement challenges (DNS, Clarity) are shown and compared.

show abstract

“…Dataset 1: The audio dataset was provided by the 1st Clarity enhancement challenge [22]. It consists of 6,000 scenes including 24 different speakers.…”

Section: Datasetsmentioning

confidence: 99%

An End-to-End Deep Learning Speech Coding and Denoising Strategy for Cochlear Implants

Gajęcki

Nogueira

2021

Preprint

View full text Add to dashboard Cite

Cochlear implant (CI) users struggle to understand speech in noisy conditions. In this work, we propose an end-to-end speech coding and denoising sound coding strategy that estimates the electrodograms from the raw audio captured by the microphone. We compared this approach to a classic Wiener filter and TasNet to assess its potential benefits in the context of electric hearing. The performance of the network is assessed by means of noise reduction performance (signal-to-noise-ratio improvement) and objective speech intelligibility measures. Furthermore, speech intelligibility was measured in 5 CI users to assess the potential benefits of each of the investigated algorithms. Results suggest that the speech performance of the tested group seemed to be equally good using our method compared to the front-end speech enhancement algorithm.

show abstract

Clarity-2021 Challenges: Machine Learning Challenges for Advancing Hearing Aid Processing

Cited by 48 publications

References 18 publications

Deep Latent Fusion Layers for Binaural Speech Enhancement

Deep Latent Fusion Layers for Binaural Speech Enhancement

A Survey on Low-Latency DNN-Based Speech Enhancement

An End-to-End Deep Learning Speech Coding and Denoising Strategy for Cochlear Implants

Contact Info

Product

Resources

About