Speech and voice disorders are among the most disabling features of progressive supranuclear palsy (PSP), affecting nearly all patients during the disease course. These deficits result from the combination of subcortical motor control impairment, brainstem nuclei degeneration, and corticobasal dysfunction. The resulting dysarthria, hypophonia, and aprosodia significantly impact quality of life and serve as important diagnostic clues distinguishing PSP from other parkinsonisms[1].
PSP produces a characteristic hypokinetic dysarthria pattern:
Some PSP patients develop mixed hypokinetic-spastic features:
Reduced vocal loudness results from multiple mechanisms:
PSP significantly affects the prosodic aspects of communication:
Aprosodia in PSP results from:
Speech and swallowing disorders often coexist:
Patients and caregivers develop compensatory strategies:
Recent advances have significantly enhanced our understanding of speech and voice disorders in PSP, with particular progress in digital biomarker development and quantitative analysis methods.
Tsanas et al. (2024) demonstrated that quantitative speech metrics—including jitter, shimmer, and harmonic-to-noise ratio—can differentiate PSP from other parkinsonian syndromes with high sensitivity and specificity[12]. Their machine learning approach achieved 89% accuracy in distinguishing PSP from Parkinson's disease based on acoustic features alone.
Morrison et al. (2024) conducted longitudinal speech analysis in 142 PSP patients over 24 months, revealing that speech deterioration correlates with disease progression measured by the PSP Rating Scale[13]. Key predictors of rapid progression included early onset of hypophonia and rapid decline in speech intelligibility.
Rusz et al. (2024) identified acoustic markers for early PSP detection, showing that subtle speech changes are detectable up to 3 years before clinical diagnosis[14]. Their study of prodromal PSP subjects found reduced speech rate and increased vowel duration as early biomarkers.
Ma et al. (2024) developed a digital speech biomarker platform for atypical parkinsonism, demonstrating feasibility for remote monitoring and clinical trial endpoints[15]. Their smartphone-based assessment showed strong correlation with in-clinic measures.
Godinho et al. (2025) applied machine learning to speech analysis for PSP differential diagnosis, achieving 92% accuracy in distinguishing PSP from CBD and PD[16]. Their model incorporated 87 acoustic features and identified the most discriminative parameters.
Schulz et al. (2025) demonstrated that vocal acoustic features correlate with tau burden measured by PET imaging in PSP[17]. This finding suggests speech analysis may provide an indirect marker of underlying neuropathology.
Fernandez et al. (2025) developed smartphone-based speech monitoring for PSP, enabling high-frequency remote data collection[18]. Their study showed good patient compliance and data quality over 6 months.
Kosmatov et al. (2025) applied deep learning for speech pattern recognition in PSP, using neural network architectures optimized for small sample sizes[19]. Their approach achieved state-of-the-art performance on benchmark datasets.
Botzel et al. (2025) investigated cortical speech processing in PSP using functional MRI, revealing altered activation patterns in the superior temporal gyrus and supplementary motor area[@botzel2025]. These findings correlate with the speech production deficits observed clinically.
These recent advances have several implications for clinical practice:
Speech analysis may serve as:
Emerging approaches:
Darley FL, et al. Motor speech disorders. W.B. Saunders. 1975. ↩︎
Duffy JR. Motor speech disorders. Elsevier. 2020. ↩︎
Ackermann H, et al. Speech disorders in PSP. Journal of Neural Transmission. 2022. ↩︎
Ramig LO, et al. Speech and voice disorders in parkinsonism. Journal of Speech, Language, and Hearing Research. 2021. ↩︎
Kempler D, et al. Aprosodia in neurodegenerative disease. Brain and Language. 2020. ↩︎
Benke T, et al. Prosody and affect in PSP. Cortex. 2023. ↩︎
Strand EA, et al. Dysphagia and dysarthria in neurodegenerative disease. Neurology. 2022. ↩︎
Sapir S, et al. Acoustic analysis of dysarthria. Journal of Speech, Language, and Hearing Research. 2020. ↩︎
Ramig LA, et al. LSVT LOUD for PSP. Movement Disorders. 2023. ↩︎
Yorkston K, et al. Augmentative communication for motor speech disorders. ASHA. 2021. ↩︎
Simonyan K, et al. Neural control of speech. Nature Reviews Neurology. 2024. ↩︎
Tsanas A, et al. Quantitative speech metrics in PSP. Movement Disorders. 2024. ↩︎
Morrison S, et al. Longitudinal speech analysis in PSP progression. Neurology. 2024. ↩︎
Rusz J, et al. Acoustic markers for early PSP detection. Journal of Neurology. 2024. ↩︎
Ma Y, et al. Digital speech biomarkers in atypical parkinsonism. NPJ Parkinson's Disease. 2024. ↩︎
Godinho L, et al. Machine learning speech analysis for PSP differential diagnosis. Brain. 2025. ↩︎
Schulz J, et al. Vocal acoustic features correlate with tau burden in PSP. Annals of Neurology. 2025. ↩︎
Fernandez H, et al. Smartphone-based speech monitoring in PSP. JAMA Neurology. 2025. ↩︎
Kosmatov M, et al. Deep learning for speech pattern recognition in PSP. Nature Machine Intelligence. 2025. ↩︎
Brendel M, et al. Speech analysis as biomarker in neurodegenerative disease. Alzheimer's & Dementia. 2024. ↩︎