Videoconferencing is now a frequent mode of communication in both professional and informal settings, yet it often lacks the fluidity and enjoyment of in-person conversation. This study leverages multimodal machine learning to predict moments of negative experience in videoconferencing. We sampled thousands of short clips from the RoomReader corpus [1], extracting audio embeddings, facial actions, and body motion features to train models for identifying low conversational fluidity, low enjoyment, and classifying conversational events (backchanneling, interruption, or gap). Our best models achieved an ROC-AUC of up to 0.87 on hold-out videoconference sessions, with domain-general audio features proving most critical. This work demonstrates that multimodal audio-video signals can effectively predict high-level subjective conversational outcomes. In addition, this is a contribution to research on videoconferencing user experience by showing that multimodal machine learning can be used to identify rare moments of negative user experience for further study or mitigation.

Audio DNNs have demonstrated impressive performance on various machine listening tasks; however, most of their representations are computationally costly and uninterpretable, leaving room for optimization. Here, we propose a novel approach centered on spectrotemporal modulation (STM) features, a signal processing method that mimics the neurophysiological representation in the human auditory cortex. The classification performance of our STM-based model, without any pretraining, is comparable to that of pretrained audio DNNs across diverse naturalistic speech, music, and environmental sounds, which are essential categories for both human cognition and machine perception. These results show that STM is an efficient and interpretable feature representation for audio classification, advancing the development of machine listening and unlocking exciting new possibilities for basic understanding of speech and auditory sciences, as well as developing audio BCI and cognitive computing.

Group conversations over videoconferencing are a complex social behavior. However, the subjective moments of negative experience, where the conversation loses fluidity or enjoyment remain understudied. These moments are infrequent in naturalistic data, and thus training a supervised learning (SL) model requires costly manual data annotation. We applied semi-supervised learning (SSL) to leverage targeted labeled and unlabeled clips for training multimodal (audio, facial, text) deep features to predict non-fluid or unenjoyable moments in holdout videoconference sessions. The modality-fused co-training SSL achieved an ROC-AUC of 0.9 and an F1 score of 0.6, outperforming SL models by up to 4% with the same amount of labeled data. Remarkably, the best SSL model with just 8% labeled data matched 96% of the SL model's full-data performance. This shows an annotation-efficient framework for modeling videoconference experience.

Inner Interpretability is a promising emerging field tasked with uncovering the inner mechanisms of AI systems, though how to develop these mechanistic theories is still much debated. Moreover, recent critiques raise issues that question its usefulness to advance the broader goals of AI. However, it has been overlooked that these issues resemble those that have been grappled with in another field: Cognitive Neuroscience. Here we draw the relevant connections and highlight lessons that can be transferred productively between fields. Based on these, we propose a general conceptual framework and give concrete methodological strategies for building mechanistic explanations in AI inner interpretability research. With this conceptual framework, Inner Interpretability can fend off critiques and position itself on a productive path to explain AI systems.

Natural and artificial audition can in principle acquire different solutions to a given problem. The constraints of the task, however, can nudge the cognitive science and engineering of audition to qualitatively converge, suggesting that a closer mutual examination would potentially enrich artificial hearing systems and process models of the mind and brain. Speech recognition - an area ripe for such exploration - is inherently robust in humans to a number transformations at various spectrotemporal granularities. To what extent are these robustness profiles accounted for by high-performing neural network systems? We bring together experiments in speech recognition under a single synthesis framework to evaluate state-of-the-art neural networks as stimulus-computable, optimized observers. In a series of experiments, we (1) clarify how influential speech manipulations in the literature relate to each other and to natural speech, (2) show the granularities at which machines exhibit out-of-distribution robustness, reproducing classical perceptual phenomena in humans, (3) identify the specific conditions where model predictions of human performance differ, and (4) demonstrate a crucial failure of all artificial systems to perceptually recover where humans do, suggesting alternative directions for theory and model building. These findings encourage a tighter synergy between the cognitive science and engineering of audition.

We present a new feature representation for speech recognition based on both amplitude modulation spectra (AMS) and frequency modulation spectra (FMS). A comprehensive modulation spectral (CMS) approach is defined and analyzed based on a modulation model of the band-pass signal. The speech signal is processed first by a bank of specially designed auditory band-pass filters. CMS are extracted from the output of the filters as the features for automatic speech recognition (ASR). A significant improvement is demonstrated in performance on noisy speech. On the Aurora 2 task the new features result in an improvement of 23.43% relative to traditional mel-cepstrum front-end features using a 3 GMM HMM back-end. Although the improvements are relatively modest, the novelty of the method and its potential for performance enhancement warrants serious attention for future-generation ASR applications.

We have developed a method suitable for reconstructing spatio-temporal activities of neural sources using MEG data. Our method is based on an adaptive beamformer technique. It extends a beamformer originally proposed by Borgiotti and Kaplan (1979) to a vector beamformer formulation in which three sets of weight vectors are used to detect the source activity in three orthogonal directions. The weight vectors of this vector-extension of the Borgiotti-Kaplan beamformer are then projected onto the signal subspace of the measurement covariance matrix to obtain a final form of the proposed beamformer's weight vectors. Our numerical experiments demonstrated the effectiveness of the proposed beamformer.

We propose a method that incorporates the time-frequency characteristics of neural sources into magnetoencephalographic (MEG) source estimation. The method is based on the multiple-signal-classification (MUSIC) algorithm and it calculates a time--frequency matrix in which diagonal and off-diagonal terms are the auto and crosstime--frequency distributions of multichannel MEG recordings, respectively. The method averages this time-frequency matrix over the time--frequency region of interest. The locations of neural sources are then estimated by checking the orthogonality between the noise subspace of this averaged matrix and the sensor lead field. Accordingly, the method allows us to estimate the locations of neural sources from each time--frequency component. A computer simulation was performed to test the validity of the proposed method, and the results demonstrate its effectiveness.