Multiple Sclerosis is a chronic inflammatory disease of the central nervous system. Over time, people with MS may experience significant changes in cognition, language and speech processes. In this study we investigate speech utterances recorded over the course of three years for 16 MS subjects and 12 healthy controls. Our examination is based on speaker category classification (healthy or MS) using wav2vec2 embeddings as features. We found that subject classification performance improved over time: the 0.745-0.844 AUC values from year one increased to 0.891-0.979 in the third year. By analyzing the posterior estimates, we measured a statistically significant improvement in the scores corresponding to the third year for the MS category, while for the control subjects there was no such tendency. This, in our view, indicates that the change is due to a subtle deterioration in the condition of MS patients, which was detected by our machine learning workflow.

In the past few years, self-supervised learning has revolutionalized automatic speech recognition.Self-supervised models such as wav2vec2, due to their generalization ability on huge unannotated audio corpora, were claimed to be state-ofthe-art feature extractors in paralinguistic and pathological applications as well.In this study we test embeddings extracted from a wav2vec 2.0 model fine-tuned on the target language as features on a multiple sclerosis audio corpus, using three speech tasks.After comparing the resulting classification performances with traditional features such as ComParE functionals, ECAPA-TDNN and activations of a HMM/DNN hybrid acoustic model, we found that wav2vec2-based models, surprisingly, only produced a mediocre classification performance.In contrast, the decade-old ComParE functionals feature set consistently led to high scores.Our results also indicate that the number of features correlates surprisingly well with classification performance.

Multiple sclerosis (MS) is a chronic inflammatory disease of the central nervous system. It affects cognitive and motor functions, and the limitation of executive functions can also manifest itself in speech production. Due to this, automatic speech analysis might serve as an effective technique for assessing MS, or for monitoring the status of the patient. However, choosing the features to be extracted from the recordings is not straightforward. In the past few years, general feature extractors such as i-vectors, d-vectors and x-vectors have found their way into automatic speech analysis. In this study we show that there is no need to employ a special neural network architecture such as x-vectors to calculate effective features, but (even more) indicative features can be derived on the basis of a standard Deep Neural Network acoustic model. From our results, these features could effectively be used to distinguish MS subjects from healthy controls, as we measured AUC scores up to 0.935. We found that classification performance depended only slightly on the choice of the hid-den layer used to extract our features, but the speech task per-formed by the subject turned out to be an important factor.

Dementia is a chronic or progressive clinical syndrome, mainly characterized by the deterioration of memory, thinking, reasoning and language. In Mild cognitive impairment (MCI), often considered as the prodromal stage of dementia, there is also a subtle deterioration of these functions, but they do not affect the daily life of the patient. However, due to the slight nature of the changes, it is quite hard to diagnose MCI. In this study, we employ sequence-to-sequence deep autoencoders in order to extract compact, robust and efficient attributes from the spontaneous speech of 25 MCI subjects and 25 healthy controls. From our results, this approach gives a competitive performance, as we significantly outperformed x-vectors even though they were trained on more data. Our additional efforts to identify mild Alzheimer’s (mAD) subjects as well were less successful; but since the focus is on the early detection of dementia, this is not a limitation of the methodology from a practical point of view.

Abstract In this article, we seek to automatically identify Hungarian patients suffering from mild cognitive impairment (MCI) or mild Alzheimer disease (mAD) based on their speech transcripts, focusing only on linguistic features. In addition to the features examined in our earlier study, we introduce syntactic, semantic, and pragmatic features of spontaneous speech that might affect the detection of dementia. In order to ascertain the most useful features for distinguishing healthy controls, MCI patients, and mAD patients, we carry out a statistical analysis of the data and investigate the significance level of the extracted features among various speaker group pairs and for various speaking tasks. In the second part of the article, we use this rich feature set as a basis for an effective discrimination among the three speaker groups. In our machine learning experiments, we analyze the efficacy of each feature group separately. Our model that uses all the features achieves competitive scores, either with or without demographic information (3-class accuracy values: 68%–70%, 2-class accuracy values: 77.3%–80%). We also analyze how different data recording scenarios affect linguistic features and how they can be productively used when distinguishing MCI patients from healthy controls.

Schizophrenia is a heterogeneous chronic and severe mental disorder.There are several different theories for the development of schizophrenia from an etiological point of view: neurochemical, neuroanatomical, psychological and genetic factors may also be present in the background of the disease.In this study, we examined spontaneous speech productions by patients suffering from schizophrenia (SCH) and bipolar disorder (BD).We extracted 15 temporal parameters from the speech excerpts and used machine learning techniques for distinguishing the SCH and BD groups, their subgroups (SCH-S and SCH-Z) and subtypes (BD-I and BD-II).Our results indicated, that there is a notable difference between spontaneous speech productions of certain subgroups, while some appears to be indistinguishable for the used classification model.Firstly, SCH and BD groups were found to be different.Secondly, the results of SCH-S subgroup were distinct from BD. Thirdly, the spontaneous speech of the SCH-Z subgroup was found to be very similar to the BD-I, however, it was sharply distinct from BD-II.Our detailed examination highlighted the indistinguishable subgroups and led to us to make our S and Z theory more clarified.

Schizophrenia is a neurodegenerative disease with spectrum disorder, consisting of groups of different deficits.It is, among other symptoms, characterized by reduced information processing speed and deficits in verbal fluency.In this study we focus on the speech production fluency of patients with schizophrenia compared to healthy controls.Our aim is to show that a temporal speech parameter set consisting of articulation tempo, speech tempo and various pause-related indicators, originally defined for the sake of early detection of various dementia types such as Mild Cognitive Impairment and early Alzheimer's Disease, is able to capture specific differences in the spontaneous speech of the two groups.We tested the applicability of the temporal indicators by machine learning (i.e. by using Support-Vector Machines).Our results show that members of the two speaker groups could be identified with classification accuracy scores of between 70 -80% and F-measure scores between 81% and 87%.Our detailed examination revealed that, among the pause-related temporal parameters, the most useful for distinguishing the two speaker groups were those which took into account both the silent and filled pauses.

Mild Cognitive Impairment (MCI), sometimes regarded as a prodromal stage of Alzheimer's disease, is a mental disorder that is difficult to diagnose.Recent studies reported that MCI causes slight changes in the speech of the patient.Our previous studies showed that MCI can be efficiently classified by machine learning methods such as Support-Vector Machines and Random Forest, using features describing the amount of pause in the spontaneous speech of the subject.Furthermore, as hesitation is the most important indicator of MCI, we took special care when handling filled pauses, which usually correspond to hesitation.In contrast to our previous studies which employed manually constructed feature sets, we now employ (automatic) correlation-based feature selection methods to find the relevant feature subset for MCI classification.By analyzing the selected feature subsets we also show that features related to filled pauses are useful for MCI detection from speech samples.

Mild Cognitive Impairment (MCI), sometimes regarded as a prodromal stage of Alzheimer's disease, is a mental disorder that is difficult to diagnose.However, recent studies reported that MCI causes slight changes in the speech of the patient.Our starting point here is a study that found acoustic correlates of MCI, but extracted the proposed features manually.Here, we automate the extraction of the features by applying automatic speech recognition (ASR).Unlike earlier authors, we use ASR to extract only a phonetic level segmentation and annotation.While the phonetic output allows the calculation of features like the speech rate, it avoids the problems caused by the agrammatical speech frequently produced by the targeted patient group.Furthermore, as hesitation is the most important indicator of MCI, we take special care when handling filled pauses, which usually correspond to hesitation.Using the ASR-based features, we employ machine learning methods to separate the subjects with MCI from the control group.The classification results obtained with ASR-based feature extraction are just slightly worse that those got with the manual method.The F1 value achieved (85.3) is very promising regarding the creation of an automated MCI screening application.