Analysis of acoustic space variability in speech affected by depression

Abstract

The spectral and energy properties of speech have consistently been observed to change with a speaker's level of clinical depression. This has resulted in spectral and energy based features being a key component in many speech-based classification and prediction systems. However there has been no in-depth investigation into understanding how acoustic models of spectral features are affected by depression. This paper investigates the hypothesis that the effects of depression in speech manifest as a reduction in the spread of phonetic events in acoustic space as modelled by Gaussian Mixture Models (GMM) in combination with Mel Frequency Cepstral Coefficients (MFCC).Our investigation uses three measures of acoustic variability: Average Weighted Variance (AWV), Acoustic Movement (AM) and Acoustic Volume, which attempt to model depression specific acoustic variations (AWV and Acoustic Volume), or the trajectory of a speech in the acoustic space (AM).Within our analysis we present the Probabilistic Acoustic Volume (PAV) a novel method for robustly estimating Acoustic Volume using a Monte Carlo sampling of the feature distribution being * Corresponding Author modelled. We show that using an array of PAV points we gain insights into how the concentration of the feature vectors in the feature space changes with depression. Key results -found on two commonly used depression corpora -consistently indicate that as a speaker's level of depression increases there are statistically significantly reductions in both AWV (-0.44 ≤ r s ≤ -0.18 with p < .05) and AM (-0.26 ≤ r s ≤ -0.19 with p < .05) values, indicating a decrease in localised acoustic variance and smoothing in acoustic trajectory respectively. Further there are also statistically significant reductions (-0.32 ≤ r s ≤ -0.20 with p < .05) in Acoustic Volume measures and strong statistical evidence (-0.48 ≤ r s ≤ -0.23 with p < .05) that the MFCC feature space becomes more concentrated.Quantifying these effects is expected to be a key step towards building an objective classification or prediction system which is robust to many of the unwanted -in terms of depression analysissources of variability modulated into a speech signal.

Implementing Gender-Dependent Vowel-Level Analysis for Boosting Speech-Based Depression Recognition

“…They outperformed, a range of state-of-the-art approaches including Vocal Tract Correlation features, i-vectors, and a deep neural network. Our results confirm two key findings presented in the literature: firstly, depression manifests at the phoneme level of speech [21]; and secondly, the effects of depression in speech can be captured by features which characterise speech motor control [7,8].…”

“…Very recent research suggest that depression impacts speech motor control [7,8]. Depression, similar to many speech motor control disorders [9], can be characterised by prosodic abnormalities, articulatory and phonetic errors [3].…”

“…This work builds on previous results offered in [21], which demonstrate the potential of using phoneme based features for depression detection. Further, by analysis of the formant features, we also aim to capture effects relating to depression-induced changes in speech motor control [7,8], to aid the classification of depression affected speech. Earlier phoneme-level analysis provided an outstanding classification performance for cross-corpus emotion recognition [22,23].…”

Dynamic Multimodal Measurement of Depression Severity Using Deep Autoencoding

“…The latter use signal processing, computer vision, and pattern recognition methodologies. From the computer-science perspective, research has sought to identify depression from vocal utterances [8], [9], [10], [11], [12], [13], facial expression [14], [15], [16], [17], head movements/pose [18], [16], [19], body movements [18], and gaze [20]. While most research is limited to a single modality, there is increasing interest in multimodal approaches to depression detection [21], [22].…”

Detecting Depression in Speech: A Multi-classifier System with Ensemble Pruning on Kappa-Error Diagram

“…With the development of machine learning and affective sensing technology, many approaches on the automatic detection of depression using speech signals have been investigated lately [25,26]. A wide range of features have been explored for the automatic classification of depressed speech.…”