Pitch-Synchronous Voice Biomarkers for Enhanced Disease Detection
This technology is a pitch-synchronous voice analysis method that separates timbre from pitch to generate more precise vocal biomarkers for non-invasive disease screening and monitoring.
Unmet Need: Reliable voice biomarkers that preserve pitch and timbre information
Voice-based diagnostic methods commonly rely on acoustic features such as Mel-frequency cepstral coefficients (MFCCs), which are calculated over fixed time windows. This approach can mix pitch and timbre information and miss subtle changes within individual voice cycles, reducing the sensitivity and reproducibility of resulting biomarkers. More precise acoustic features are needed to support reliable comparison across voice samples and improve the performance of diagnostic models.
The Technology: Pitch-synchronous timbre analysis for reliable voice biomarkers
This technology is a pitch-synchronous voice analysis method that analyzes voice signals in alignment with individual pitch periods rather than using fixed time windows. It extracts pitch-synchronous timbre vectors and instantaneous pitch information, allowing pitch and timbre characteristics to be evaluated separately while preserving fine-grained acoustic variation. The resulting features can be compared using a timbre-distance metric and used as biomarkers in voice-based screening, monitoring, and classification models.
Demonstrated in silico with simulations and initial analyses showing enhanced feature extraction compared to traditional MFCCs.
Applications:
- Voice-based diagnostics for neurological, psychiatric, and respiratory conditions
- Remote patient monitoring and telemedicine platforms
- Digital therapeutics and mental health monitoring tools
- Clinical decision-support and diagnostic software development
- Voice authentication and biometric security systems
- Research tools for voice science, speech disorders, and otolaryngology
Advantages:
- Separates pitch and timbre information for more precise voice characterization
- Preserves fine-grained acoustic features that fixed-window methods may overlook
- Provides a quantitative timbre-distance metric for robust voice comparison
- Improves the sensitivity and reproducibility of voice-based biomarkers
- Generates features that can be integrated into diagnostic and classification models
- Supports applications across clinical diagnostics, monitoring, research, and biometrics
Lead Inventor:
Patent Information:
Patent Pending
Related Publications:
Tech Ventures Reference:
IR CU26053
Licensing Contact: Dovina Qu