William Demant Holding (Denmark)

Advanced auditory models are useful in designing signal-processing algorithms for hearing-loss compensation or speech enhancement. Such auditory models provide rich and detailed descriptions of the auditory pathway, and might allow for individualization of signal-processing strategies, based on physiological measurements. However, these auditory models are often computationally demanding, requiring significant time to compute. To address this issue, previous studies have explored the use of deep neural networks to emulate auditory models and reduce inference time. While these deep neural networks offer impressive efficiency gains in terms of computational time, they may suffer from uneven emulation performance as a function of auditory-model frequency-channels and input sound pressure level, making them unsuitable for many tasks. In this study, we demonstrate that the conventional machine-learning optimization objective used in existing state-of-the-art methods is the primary source of this limitation. Specifically, the optimization objective fails to account for the frequency- and level-dependencies of the auditory model, caused by a large input dynamic range and different types of hearing losses emulated by the auditory model. To overcome this limitation, we propose a new optimization objective that explicitly embeds the frequency- and level-dependencies of the auditory model. Our results show that this new optimization objective significantly improves the emulation performance of deep neural networks across relevant input sound levels and auditory-model frequency channels, without increasing the computational load during inference. Addressing these limitations is essential for advancing the application of auditory models in signal-processing tasks, ensuring their efficacy in diverse scenarios.

In daily life, both acoustic factors and social context can affect listening effort investment. In laboratory settings, information about listening effort has been deduced from pupil and cardiovascular responses independently. The extent to which these measures can jointly predict listening-related factors is unknown. Here we combined pupil and cardiovascular features to predict acoustic and contextual aspects of speech perception. Data were collected from 29 adults (mean = 64.6 years, SD = 9.2) with hearing loss. Participants performed a speech perception task at two individualized signal-to-noise ratios (corresponding to 50% and 80% of sentences correct) and in two social contexts (the presence and absence of two observers). Seven features were extracted per trial: baseline pupil size, peak pupil dilation, mean pupil dilation, interbeat interval, blood volume pulse amplitude, pre-ejection period and pulse arrival time. These features were used to train k-nearest neighbor classifiers to predict task demand, social context and sentence accuracy. The k-fold cross validation on the group-level data revealed above-chance classification accuracies: task demand, 64.4%; social context, 78.3%; and sentence accuracy, 55.1%. However, classification accuracies diminished when the classifiers were trained and tested on data from different participants. Individually trained classifiers (one per participant) performed better than group-level classifiers: 71.7% (SD = 10.2) for task demand, 88.0% (SD = 7.5) for social context, and 60.0% (SD = 13.1) for sentence accuracy. We demonstrated that classifiers trained on group-level physiological data to predict aspects of speech perception generalized poorly to novel participants. Individually calibrated classifiers hold more promise for future applications.

Humans can listen to a target speaker even in challenging acoustic conditions that have noise, reverberation, and interfering speakers. This phenomenon is known as the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">cocktail party effect</i> . For decades, researchers have focused on approaching the listening ability of humans. One critical issue is handling interfering speakers because the target and nontarget speech signals share similar characteristics, complicating their discrimination. Target speech/speaker extraction (TSE) isolates the speech signal of a target speaker from a mixture of several speakers, with or without noises and reverberations, using clues that identify the speaker in the mixture. Such clues might be a spatial clue indicating the direction of the target speaker, a video of the speaker's lips, and a prerecorded enrollment utterance from which the speaker's voice characteristics can be derived. TSE is an emerging field of research that has received increased attention in recent years because it offers a practical approach to the cocktail party problem and involves such aspects of signal processing as audio, visual, and array processing as well as deep learning. This article focuses on recent neural-based approaches and presents an in-depth overview of TSE. We guide readers through the different major approaches, emphasizing the similarities among frameworks and discussing potential future directions.

Acoustic beamforming is crucial for many applications where ex-traction of a target signal from a noisy environment is required. In order to implement practical beamformers, e.g. the multichannel Wiener filter (MWF), estimation of the target and noise power spectral densities (PSDs), and the relative acoustic transfer functions (RATFs) is essential. Several methods, e.g. the so-called covariance whitening (CW) approach, have been proposed for estimating these parameters. However, it seems largely unknown that the CW approach in fact leads to maximum likelihood (ML) estimates of the RATFs. We use historical results to derive joint ML estimates (MLEs) of the RATFs and PSDs in the context of acoustic beam-forming. In addition, based on the MLEs, we propose a basic VAD framework using concentrated likelihood ratios. We use the joint MLEs of the PSDs, RATFs, and the proposed VAD to implement beamformers in a hearing aid application, and compare its performance to competing methods. Simulation results show that the pro-posed scheme can outperform competing methods, in particular in realistic situations where highly accurate prior RATF knowledge is not available or at higher signal-to-noise ratios.

This work addresses the needs companies face when assessing and tracking the UX quality of their products, which is necessary to ensure that the desired level of quality is met, clarify potential areas of improvement, and compare competitor products via benchmarking.

This chapter examines students’ perception of the PBL environment. It is written by two graduate students who studied in a PBL environment. Their PBL experiences were gained within the field of International Business Economics. To understand their perceptions information is provided on their cultural background, previous study-related experiences and current work position. They identify and compare the key competences which were the objectives of the teacher with those that they acquired during their study and compare these with the competences which they find they need in their employment. They reflect on the extent to which their experience in learning, being evaluated and assessed in a PBL environment, focusing of their project work, group work, independence in problem formulation, overall freedom and degree in their studies to choose (focus on self-directed learning) has equipped them for their employment.