Prosodic features in Finnish-speaking adults with Parkinson´s disease

Figures & data

Table 1. Prosodic variables (see, Bunton et al., 2001; Daniele & Patel, 2013; Grabe & Low, 2002; Y. Kim & Choi, 2017; Tavi & Werner, 2020; Tsao et al., 2006).

Variable	Definition
Pitch	average fundamental frequency (F0) shows the typical pitch standard deviation of F0 indicates pitch variability can be used as a measure of intonation
Articulation rate	number of syllables a person can produce in a unit of time excluding pauses usually expressed as syllables per second (syl/sec) can be used as a measure of rhythm
Energy proportion below 1 kHz (epb1kHz)	percentage of frequencies in the voice frequency spectrum below 1 kHz $epb1kHz=\frac{bandofenergybetween0-1kHz}{bandofenergybetween0-4kHz}$ can be used as a measure of prosodic emphasis and voice quality
Normalised pairwise variability index (nPVI)	equation that measures how much durational contrast exists between successive events in an utterance $\mathrm{n}PVI=\frac{100}{m-1}\times \sum _{k=1}^{m-1}\left|\frac{d_k-d_{k+1}}{\frac{d_k+d_{k+1}}{2}}\right|$ can be used as a measure of rhythm
Syllabic prosody index (SPI)	combines median pitch, energy balance, and duration in syllables $SPI=\frac{Pitc{h}_{median}\times \sqrt{Duration}}{\sqrt{Energ{y}_{below1kHz}}}/10.$ can be used as a measure of prosodic prominence

Table 2. Participant groups.

		Speakers with Parkinson’s disease			Control speakers
		All (n = 16)	Female (n = 8)	Male (n = 8)	All (n = 20)	Female (n = 10)	Male (n = 10)
Age (years)	M	69.6	65.7	73.6	62.8	60.7	65.0
	SD	7.25	7.22	5.01	7.47	3.40	9.79
Voice Handicap Index-9	M	13.9	11.3	16.5	2.9	3.0	2.7
	SD	5.6	6.4	3.2	3.4	3.7	3.3
Speech impairment (VAS 0–10)	M	2.1	1.5	2.7
	SD	1.5	0.7	1.9
Time from the diagnosis (years)	M	6.3	6.5	6.1
	SD	3.70	4.56	2.90
Hoehn and Yahr severity rating*	M	1.5	1.43	1.56
	SD	0.54	0.63	0.67

*Hoehn and Yahr scale: 1 = unilateral involvement only, 1.5 = unilateral and axial involvement, 2 = bilateral involvement without impairment of balance, 2.5 = mild bilateral disease, 3 = mild to moderate bilateral disease, 4 = severe disability, 5 = patient confined to bed or wheelchair unless aided

Figure 1. Energy distributions (energy proportion below 1 kHz) in two speaker groups.

Table 3. Prosody features in two speaker groups.

		Articulation rate	nPVI	epb1kHz	F0	F0 SD	SPI
Control group (n = 20)	Mean	5.580	41.257	0.959	142.294	27.390	6.031
	Std. Deviation	0.566	3.523	0.033	44.037	13.998	1.762
	Median	5.575	41.430	0.966	143.625	23.080	5.771
	Minimum	4.610	33.590	0.890	65.140	10.800	3.456
	Maximum	7.020	47.070	0.996	206.260	51.650	8.708
PD group(n = 16)	Mean	5.349	39.841	0.984	146.327	22.133	6.222
	Std. Deviation	0.508	3.835	0.011	25.726	6.457	1.092
	Median	5.415	40.480	0.985	139.285	21.430	6.197
	Minimum	4.170	33.990	0.950	111.950	14.580	4.902
	Maximum	6.200	47.140	0.998	185.370	38.130	8.473
Statistical significance	Level p < 0.05	ns	ns	p = 0.016	ns	ns	ns

* ns = not statistically significant

Figure 2. Typical pitch (F₀) in male speakers.

Figure 3. Syllabic prosody index (SPI) in male speakers.

Table 4. Prosody features in subgroups.

		Articulation rate	nPVI	epb1kHz	F0	F0 SD	SPI
PD males (n = 8)	Mean	5.266	39.048	0.990	123.491	18.692	5.277
	Std. Deviation	0.569	5.155	0.005	7.830	4.827	0.344
	Median	5.205	37.005	0.990	122.340	16.785	5.233
	Minimum	4.170	33.990	0.983	111.950	14.580	4.902
	Maximum	6.140	47.140	0.998	134.450	29.130	6.002
Control males (n = 10)	Mean	5.781	40.829	0.959	103.990	16.913	4.502
	Std. Deviation	0.665	3.910	0.036	21.137	4.628	0.783
	Median	5.650	40.680	0.964	104.545	15.965	4.390
	Minimum	4.980	33.590	0.890	65.140	10.800	3.456
	Maximum	7.020	47.070	0.996	143.590	23.620	5.826
Statistical significance	Level p < 0.05	ns	ns	ns	p = 0.013	ns	p = 0.026
PD females (n = 8)	Mean	5.432	40.633	0.978	169.163	25.573	7.168
	Std. Deviation	0.465	1.872	0.013	12.844	6.245	0.628
	Median	5.495	40.625	0.983	169.510	23.980	7.115
	Minimum	4.630	37.420	0.950	144.120	18.750	6.392
	Maximum	6.200	43.660	0.988	185.370	38.130	8.473
Control females (n = 10)	Mean	5.380	41.685	0.959	180.598	37.867	7.560
	Std. Deviation	0.381	3.241	0.031	19.669	12.178	0.867
	Median	5.550	41.840	0.969	181.625	39.300	7.572
	Minimum	4.610	34.560	0.902	143.660	20.020	5.716
	Maximum	5.850	46.800	0.991	206.260	51.650	8.708
Statistical significance	Level p < 0.05	ns	ns	ns	ns	p = 0.033	ns

* ns = not statistically significant

Figure 4. Pitch variability (F₀SD) in female speakers.

Bunton, K., Kent, R., Kent, J., & Duffy, J. (2001). The effects of flattening fundamental frequency contours on sentence intelligibility in speakers with dysarthria. Clinical Linguistics & Phonetics, 15(3), 181–193. https://doi.org/10.1080/02699200010003378

 Web of Science ®Google Scholar

Daniele, J., & Patel, A. (2013). An empirical study of historical patterns in musical rhythm. Analysis of German & Italian classical music using the nPVI equation. Music Perception: An Interdisciplinary Journal, 31(1), 10–18. https://doi.org/10.1525/mp.2013.31.1.10

 Web of Science ®Google Scholar

Grabe, E., & Low, E. L. (2002). Durational variability in speech and the rhythm class hypothesis. In N. Warner & C. Gussenhoven (Eds.), Papers in laboratory phonology 7 (pp. 515–546). Mouton de Gruyter.

 Google Scholar

Kim, Y., & Choi, Y. (2017). A Cross-language study of acoustic predictors of speech intelligibility in individuals with Parkinson’s Disease. Journal of Speech, Language and Hearing Research, 60(9), 2506–2518. https://doi.org/10.1044/2017_JSLHR-S-16-0121

 PubMed Web of Science ®Google Scholar

Tavi, L., & Werner, S. (2020). A phonetic case study on prosodic variability in suicidal emergency calls. International Journal of Speech Language and the Law, 27(1), 59–74. https://doi.org/10.1558/ijsll.39667

 Web of Science ®Google Scholar

Tsao, Y. C., Weismer, G., & Iqbal, K. (2006). Interspeaker variation in habitual speaking rate: Additional evidence. Journal of Speech, Language, and Hearing Research, 49(5), 1156–1164. https://doi.org/10.1044/1092-4388(2006/083)

 PubMed Web of Science ®Google Scholar