Paula Andrea Pérez-Toro

A standardized evaluation framework reveals that even near-perfect machine learning scores in power system protection can be misleading without consistent assessment criteria.

PhonoQ-derived features boost phonological classification accuracy, revealing intricate speech patterns that traditional models miss.

Ditch phoneme-based speech processing: directly predicting phonological features from speech unlocks substantial gains in multilingual and cross-lingual speech recognition.

Wav2Vec 2.0, despite boosting accuracy in speech-based cognitive impairment detection, introduces significant gender and age biases, misclassifying females and younger individuals at a higher rate.