Speech and Voice Acoustic Analysis for Corticobasal Syndrome

Overview

Overview

Speech and voice acoustic analysis represents an emerging non-invasive biomarker for the diagnosis and monitoring of corticobasal syndrome (CBS). This approach leverages quantitative analysis of speech and voice characteristics to detect subtle neurological changes that may not be apparent through clinical examination alone. Machine learning-based speech analysis has demonstrated up to 92% accuracy in distinguishing corticobasal degeneration (CBD) from progressive supranuclear palsy (PSP) and Parkinson's disease (PD)[@godinho2025].

Unlike traditional speech assessment, acoustic analysis provides objective, reproducible measures that can be collected remotely via smartphone applications, enabling continuous monitoring and early detection of disease progression. The technique is particularly valuable for CBS because speech and language deficits often appear early in the disease course, sometimes preceding motor symptoms by months to years.

Background

Clinical Rationale

Corticobasal syndrome is characterized by progressive asymmetric rigidity, bradykinesia, dystonia, myoclonus, and cortical sensory loss. However, speech and language disturbances are among the earliest and most disabling features:

• Apraxia of speech: Present in 40-50% of classic-onset CBS and up to 80-90% of speech/language-onset CBS
• Dysarthria: Hypokinetic or spastic components affecting speech production
• Reduced speech output: Speech poverty and slowed speech rate
• Aphasia: Non-fluent aphasia with agrammatic speech patterns

Advantages over Traditional Assessment

Acoustic Features

Fundamental Frequency (F0)

Fundamental frequency (F0) represents the rate of vocal fold vibration and is a primary voice characteristic. In CBS, F0 abnormalities include:

| Parameter | CBS Finding | Clinical Significance |
|-----------|--------------|----------------------|
| Mean F0 | Reduced (hypophonia) | Indicates vocal fold adduction weakness |
| F0 variability | Decreased | Reflects reduced intonation range |
| F0 tremor | Increased 4-6 Hz oscillation | Associated with parkinsonian features |
| Maximum F0 range | Reduced | Limited pitch variation |

Studies in PD have shown F0 variability reduction of 30-50% compared to healthy controls[@tsanas2012], and similar patterns are observed in CBS due to the hypokinetic dysarthria component.

Formants

Formants are resonant frequencies of the vocal tract that shape vowel and consonant sounds. In CBS:

Vowel Formant Analysis:

• F1 and F2 frequencies show reduced dynamic range
• Vowel space area (VSA) is compressed, similar to PD
• Formant transition duration is prolonged due to apraxia of speech

• VSA reduction correlates with speech intelligibility
• Formant Dispersion (FDisp) increases with age but is elevated in CBS
• Second formant (F2) transitions are abnormal in apraxia of speech

Jitter

Jitter measures cycle-to-cycle frequency variation in the voice signal:

| Jitter Type | Description | CBS Pattern |
|-------------|-------------|--------------|
| Jitter (local) | F0 period variation | Elevated 20-40% |
| Jitter (rap) | Relative average perturbation | Increased |
| Jitter (ddp) | Difference of differences | Elevated |

Jitter values are typically elevated in hypokinetic dysarthria (CBS/PD) compared to healthy controls, reflecting irregular vocal fold vibration. A cutoff of >1.0% jitter has been proposed as a sensitive marker for parkinsonian speech[@silvia2012].

Shimmer

Shimmer measures cycle-to-cycle amplitude variation:

| Shimmer Type | Description | CBS Pattern |
|--------------|-------------|--------------|
| Shimmer (local) | Amplitude variation | Elevated 15-35% |
| Shimmer (dB) | Decibel variation | Increased |
| Shimmer (apq) | Average perturbation quotient | Elevated |

Shimmer is often elevated alongside jitter in parkinsonian speech, reflecting the same underlying irregular vibration of the vocal folds. The combination of elevated jitter and shimmer has been proposed as a diagnostic marker for hypokinetic dysarthria.

Harmonics-to-Noise Ratio (HNR)

HNR measures the ratio of periodic to aperiodic components in the voice:

• Normal range: 20-30 dB
• CBS pattern: Reduced to 10-15 dB
• Clinical significance: Indicates breathy/hoarse voice quality

Additional Acoustic Parameters

Temporal Measures:

• Speech rate (syllables per minute): Reduced in CBS
• Pause duration: Increased pause frequency and duration
• Voice onset time (VOT): Prolonged in apraxia of speech

• Spectral centroid: Shifted toward lower frequencies
• Spectral slope: Flattened in high-frequency range
• Mel-frequency cepstral coefficients (MFCCs): Altered pattern

Smartphone-Based Analysis Platforms

Commercial Platforms

Several commercial and research platforms now offer automated voice biomarker analysis:

| Platform | Features | Validation Status |
|----------|----------|-------------------|
| Winterlight Labs | Comprehensive speech analysis, cognitive assessment | FDA breakthrough device designation |
| ki:elements | Voice biomarkers for neurological diseases | Clinical validation ongoing |
| Aural Analytics | Real-time voice analysis, clinical-grade | Used in clinical trials |
| Sonde Health | Respiratory and voice analysis | CE marked, FDA cleared |

Implementation in CBS

Data Collection Protocol:

Smartphone Compatibility:

• iOS and Android applications available
• Built-in microphones sufficient for clinical-grade analysis
• Cloud-based processing eliminates local computational requirements

Technical Considerations

Signal Quality Requirements:

• Background noise level: <35 dB
• Sampling rate: 44.1 kHz minimum
• Bit depth: 16-bit

• Device-specific calibration may be needed
• Headphone use improves consistency
• Multiple samples per session improve reliability

Comparison with PSP and PD Speech Patterns

CBS vs Progressive Supranuclear Palsy

| Feature | CBS | PSP |
|---------|-----|-----|
| Voice quality | Hypophonic, breathy | Hypophonic, harsh |
| Speech rate | Slow, variable | Slow, monotonic |
| Articulation | Imprecise, apraxia | Imprecise |
| F0 variation | Reduced | Markedly reduced |
| Jitter | Moderately elevated | Highly elevated |
| Dysarthria type | Mixed (hypokinetic + spastic) | Predominantly hypokinetic |

Key discriminating features:

• Apraxia of speech is more prominent in CBS (80-90%) than PSP (30-40%)
• CBS shows more asymmetric speech involvement
• PSP exhibits more prominent bradykinesia affecting speech

CBS vs Parkinson's Disease

| Feature | CBS | PD |
|---------|-----|-----|
| Asymmetry | Marked, persistent | Often symmetric over time |
| F0 range | Moderately reduced | Reduced |
| Jitter/shimmer | Elevated | Elevated |
| Progression | Faster decline | Slower progression |
| Apraxia of speech | Common (40-90%) | Less common (20-30%) |

Key discriminating features:

• Sustained asymmetry in CBS vs PD
• Apraxia of speech components more severe in CBS
• CBS shows more cortical signs affecting speech

Machine Learning Differentiation

Studies using machine learning have achieved high accuracy in differentiating these disorders:

• 92% accuracy for CBD vs PSP/PD using ensemble methods (Godinho 2025)[@godinho2025]
• 88% accuracy for CBS vs PD using SVM with acoustic features
• 85% accuracy for CBS vs PSP using random forest classifiers

Machine Learning Approaches

Feature Extraction Pipeline

Raw Audio → Preprocessing → Feature Extraction → Feature Vector → ML Model
↓ ↓ ↓ ↓
Sampling Noise filtering MFCC, F0, Normalization
44.1kHz Windowing jitter, shimmer Scaling

Common feature sets:

• Praat features: Jitter, shimmer, HNR, F0, formant frequencies
• MFCCs: 13-20 coefficients capturing spectral envelope
• DeepSpeech embeddings: Neural network-based features
• i-vectors: Speaker verification techniques adapted for pathology

Classification Models

| Model | Accuracy | Advantages |
|-------|----------|------------|
| SVM | 85-90% | Works well with high-dimensional features |
| Random Forest | 88-92% | Handles feature interactions |
| CNN | 90-95% | Learns temporal patterns |
| Transformer | 92-96% | Captures long-range dependencies |

Validation and Generalization

Critical considerations for clinical deployment:

Clinical Implementation

Diagnostic Workflow

Integration with Clinical Scales

Acoustic analysis can supplement existing clinical measures:

• CBD-FRS: Correlation with speech subdomain (r=0.65-0.72)
• UPDRS speech item: Correlation with acoustic metrics (r=0.70-0.80)
• MoCA: Combined cognitive-speech assessment

Limitations

Future Directions

Cross-References

• [Speech and Language Deficits in CBS](/mechanisms/speech-language-deficits-cbs)
• [Speech/Language-Onset CBS](/diseases/speech-language-onset-cbs)
• [Wearable Accelerometry for CBS](/diagnostics/cbs-accelerometry)
• [Digital Biomarkers for Neurodegeneration](/diagnostics/digital-biomarkers)
• [LSVT Voice Therapy for CBS/PSP](/therapeutics/section-249-advanced-lsvt-voice-speech-therapy-cbs-psp)

References