Acoustic variation during passage reading for speakers with dysarthria and healthy controls

Abstract

Purpose Acoustic variation in a passage read by speakers with dysarthria and healthy speakers was examined. Method 15 speakers with Multiple Sclerosis (MS), 12 speakers with Parkinson’s disease (PD), and 14 healthy speakers were studied. Acoustic variables included measures of global speech timing (e.g., articulation rate, pause characteristics), vocal intensity (e.g., mean sound pressure level and intensity modulation), and segmental articulation (i.e., utterance-level second formant interquartile range (F2 IQR)). Acoustic measures were obtained from three segments operationally defined to represent the beginning, middle, and end of a reading passage. Two speaking conditions associated with common treatment techniques for dysarthria were included for comparison to a habitual speaking condition. These conditions included a slower-than-habitual rate (Slow) and greater-than-habitual intensity (Loud). Results There was some degree of acoustic variation across the three operationally-defined segments of the reading passage. The Slow, Loud and Habitual conditions yielded comparable characteristics of variation. Patterns of acoustic variation across the three passage segments also were largely similar across speaker groups. Conclusions Within-task acoustic variation during passage reading should be considered when making decisions regarding speech sampling in clinical practice and research. The contributions of speech disorder severity and linguistic variables to within-task acoustic change warrant further investigation. Learning outcomes Readers will be able to (1) discuss the motivation for studying and understanding within-task variation in contextual speech, (2) describe patterns of acoustic variation for speakers with dysarthria and healthy speakers during passage reading, (3) discuss the relationship between non-habitual speaking conditions and within-task variation, (4) understand the need to consider within-speaker, within-task variation in speech sampling.

Random Forest Algorithm Based on Speech for Early Identification of Parkinson’s Disease

“…To minimize the effects of dialects and special speaking styles on acoustic results, speakers who spoke with an accent or spoke in an unusual way were screened out [ 26 ]. So, thirty-six individuals with idiopathic PD (19 men and 17 women) aged 52 to 78 years (mean = 63.55, SD = 9.46) were recruited as the PD group.…”

Beyond text complexity: Production-related sources of text-based variability in oral reading fluency.

“…Moreover, because the strength of production-related effects could vary across readers, this consequence can have disproportionate effects on different groups of students. For example, this consequence would disproportionately impact students with clinical or subclinical speech motor deficits because their oral reading is especially susceptible to texts’ production constraints, and these constraints (e.g., number of syllables, intonation types) can affect clinical groups differently (see, e.g., Kuo & Tjaden, 2016; Patel et al, 2013).…”

Accurate Synthesis of Dysarthric Speech for Asr Data Augmentation

“…To synthesize dysarthric speech, there is a need to build a system controlling different characteristics of dysarthric speech for generating variant dysarthric speech. As will be discussed later in this paper, and according to a number of studies (Rudzicz, Namasivayam et al 2012, Zhang, Dang et al 2014, Bigi, Klessa et al 2015, Kuo and Tjaden 2016, Yunusova, Graham et al 2016), such a system should have the following capabilities in order to support generation of authentic and diverse speech: 1) ability to control the speaking rate (duration), pitch, energy for a variety of dysarthria severity levels, 2) ability to learn and model pause behavior of dysarthric speakers (e.g., duration of pause and pause occurrence) and control pause insertion locations and durations 3) ability to learn and model individual voice characteristics of speakers and use these to generate new speaking styles 4) ability to learn and model these characteristics from a small amount of dysarthric speech data.…”