An On-Device Speech Pipeline That Discards Audio
How to build a speech system that actually deletes audio instead of just claiming to.
Chidera Nwankwo-Reid
The Prosody Register
Features
Auditing a Vocal Biomarker Model for Demographic Bias
Models trained on narrow demographics fail silently on patients they've never learned from.
A Test-Retest Reliability Study for a Voice-Derived Score
A voice-based cognitive-load score needs test-retest reliability or it measures nothing but noise.
Wiktoria Zielinska-Brandt
More From The Prosody Register
Phrasing a Voice Score So It Stays a Wellness Claim
Claudette Marcelin