Abstract

Sinusoids are widely used to represent the oscillatory modes of musical instrument sounds in both analysis and synthesis. However, musical instrument sounds feature transients and instrumental noise that are poorly modeled with quasi-stationary sinusoids, requiring spectral decomposition and further dedicated modeling. In this work, we propose a full-band representation that fits sinusoids across the entire spectrum. We use the extended adaptive Quasi-Harmonic Model (eaQHM) to iteratively estimate amplitude- and frequency-modulated (AM–FM) sinusoids able to capture challenging features such as sharp attacks, transients, and instrumental noise. We use the signal-to-reconstruction-error ratio (SRER) as the objective measure for the analysis and synthesis of 89 musical instrument sounds from different instrumental families. We compare against quasi-stationary sinusoids and exponentially damped sinusoids. First, we show that the SRER increases with adaptation in eaQHM. Then, we show that full-band modeling with eaQHM captures partials at the higher frequency end of the spectrum that are neglected by spectral decomposition. Finally, we demonstrate that a frame size equal to three periods of the fundamental frequency results in the highest SRER with AM–FM sinusoids from eaQHM. A listening test confirmed that the musical instrument sounds resynthesized from full-band analysis with eaQHM are virtually perceptually indistinguishable from the original recordings.

Abstract

We propose a fast speech analysis method which simultaneously performs high-resolution voiced/unvoiced detection (VUD) and accurate estimation of glottal closure and glottal opening instants (GCIs and GOIs, respectively). The proposed algorithm exploits the structure of the glottal flow derivative in order to estimate GCIs and GOIs only in voiced speech using simple time-domain criteria. We compare our method with well-known GCI/GOI methods, namely, the dynamic programming projected phase-slope algorithm (DYPSA), the yet another GCI/GOI algorithm (YAGA) and the speech event detection using the residual excitation and a mean-based signal (SEDREAMS). Furthermore, we examine the performance of the aforementioned methods when combined with state-of-the-art VUD algorithms, namely, the robust algorithm for pitch tracking (RAPT) and the summation of residual harmonics (SRH). Experiments conducted on the APLAWD and SAM databases show that the proposed algorithm outperforms the state-of-the-art combinations of VUD and GCI/GOI algorithms with respect to almost all evaluation criteria for clean speech. Experiments on speech contaminated with several noise types (white Gaussian, babble, and car-interior) are also presented and discussed. The proposed algorithm outperforms the state-of-the-art combinations in most evaluation criteria for signal-to-noise ratio greater than 10 dB.

Source code is available here

Abstract

Digital cough screening for COVID-19 detection shows promise, but population differences in cough acoustics and screening accuracy require investigation. This study examined cough characteristics and COVID-19 screening performance in Lima, Peru and Montreal, Canada. Cough recordings and clinical data were prospectively collected from 605 adults. COVID-19 and other respiratory pathogens were diagnosed via NAAT. Acoustic features were extracted and compared. COVID-19 classification used eXtreme Gradient Boosting (XGBoost) and a deep learning neural network, assessed via internal and external validations for audio-only, clinical-only, and combined models. A sub-analysis explored XGBoost prediction scores by underlying disease status. Significant heterogeneity in cough acoustic features existed between Lima and Montreal cohorts. XGBoost audio-based models trained and tested in Lima showed superior performance (area under the curve [AUC]: 0.71 ± 0.08) compared to Montreal (AUC: 0.53 ± 0.04). Both models demonstrated poor external validation performance when tested on the alternate dataset. Neural network models showed similar trends. Additionally, individuals with other respiratory diseases had differing COVID-19 prediction scores between sites, suggesting epidemiological context influences model performance. Cough acoustics are population-specific, impacting cough-based classification algorithm utility across different epidemiological settings. COVID-19 cough screening models demonstrated limited transferability, highlighting challenges in developing globally applicable tools without representative training data.

Abstract

In this paper, we propose a standardized framework for automatic tuberculosis (TB) detection from cough audio and routinely collected clinical data using machine learning. While TB screening from audio has attracted growing interest, progress is difficult to measure because existing studies vary substantially in datasets, cohort definitions, feature representations, model families, validation protocols, and reported metrics. Consequently, reported gains are often not directly comparable, and it remains unclear whether improvements stem from modeling advances or from differences in data and evaluation. We address this gap by establishing a strong, well-documented baseline for TB prediction using cough recordings and accompanying clinical metadata from a recently compiled dataset from several countries. Our pipeline is reproducible end-to-end, covering feature extraction, multimodal fusion, cougher-independent evaluation, and uncertainty quantification, and it reports a consistent suite of clinically relevant metrics to enable fair comparison. We further quantify performance for cough audio-only and fused (audio + clinical metadata) models, and release the full experimental protocol to facilitate benchmarking. This baseline is intended to serve as a common reference point and to reduce methodological variance that currently holds back progress in the field.

Abstract

Background: The ability to passively and continuously monitor coughing for prolonged periods of time would significantly improve cough management and research. To date there is no automated clinically validated cough monitor that can be routinely used in clinical care and research. Here we describe the validation of such an automated cough monitor. Methods: This multicenter observational study compared the results of the Hyfe CoughMonitor wrist-worn device with manually counted coughs in subjects with a variety of etiologies as they went about their usual daily activities. We collected 24 h of continuous sounds from subjects while they simultaneously wore a CoughMonitor and an audio recorder. Coughs were labelled by multiple trained annotators who listened to the continuous audio recordings using validated methodology. The time stamps of these human-detected coughs were compared to those of the CoughMonitor to determine the system’s overall performance using event-to-event and hourly rate correlation analyses. Results: Over the 546 h monitored, 4,454 cough events were recorded; The overall sensitivity was 90.4% (95% CI of 88.3–92.2%). The overall false positive rate was 1.03 false positives per hour (95% CI of 0.84 to 1.24). The overall correlation between manual and CoughMonitor measured hourly coughing was high (Pearson correlation coefficient of 0.99). Two case studies of long-term monitoring of patients with chronic cough are presented. Conclusion: The present analysis of cough events demonstrated that the Hyfe CoughMonitor accurately reflects them with a high sensitivity and a low false positive rate. Future studies should focus on its potential role in the management of patients with cough in clinical practice.

Abstract

Speech emotion recognition (SER) has gained an increased interest during the last decades as part of enriched affective computing. As a consequence, a variety of engineering approaches have been developed addressing the challenge of the SER problem, exploiting different features, learning algorithms, and datasets. In this paper, we propose the application of the graph theory for classifying emotionally-colored speech signals. Graph theory provides tools for extracting statistical as well as structural information from any time series. We propose to use the mentioned information as a novel feature set. Furthermore, we suggest setting a unique feature-based identity for each emotion belonging to each speaker. The emotion classification is performed by a Random Forest classifier in a Leave-One-Speaker-Out Cross Validation (LOSO-CV) scheme. The proposed method is compared with two state-of-the-art approaches involving well known hand-crafted features as well as deep learning architectures operating on mel-spectrograms. Experimental results on three datasets, EMODB (German, acted) and AESDD (Greek, acted), and DEMoS (Italian, in-the-wild), reveal that our proposed method outperforms the comparative methods in these datasets. Specifically, we observe an average UAR increase of almost 18%, 8%, and 13%, respectively.

Abstract

Fricatives, and especially sibilants, are very frequently misarticulated by speakers with hearing loss. Misarticulations can result in phonemic contrast weakening or loss, compromising intelligibility. The present study focuses on the examination of acoustic characteristics of the Greek alveolar fricative /s/, an articulatorily demanding sound, produced by young adult speakers with profound hearing impairment and with normal hearing. An array of variables was examined using mixed-effects and random forest models aiming to assess the effectiveness of various measures in differentiating hearing-impaired and normal-hearing /s/ production. Significant differences were found in spectral and amplitude measures, but not in temporal measures. In hearing-impaired speech, spectral slope and RMS amplitude had significantly lower values, indicating a more distributed spectrum, suggestive of decreased flow velocity through the fricative constriction. Also, a trend for concentration of energy at lower frequencies was observed suggesting more posterior fricative articulation than normal. Moreover, measures capturing the variation of frequency and amplitude over time revealed different patterns of sibilance development across time than normal, denoting the production of a less well-formed or less sibilant /s/ by speakers with hearing impairment. The investigation of contextual effects on /s/ in hearing-impaired speech showed increased spectral variance, negative skewness and lower kurtosis in the labial (rounded) context /u/ in relation to the nonlabial contexts /i/ and /a/, indicating a more diffuse, less compact spectrum with concentration at high frequencies. Findings are discussed in relation to previous literature on fricative production by speakers with hearing impairment and normal hearing in Greek and other languages.