Dealing with Limited Speech Data and Variability Using a Hybrid Knowledge-Based and Data-Driven Approaches

Dr. Abeer Alwan of the Electrical and Computer Engineering Dept. at UCLA


3 to 4:30 p.m., Feb. 23, 2024
Passcode: LingCo1
Dealing with Limited Speech Data and Variability Using a hybrid Knowledge-
Based and Data-Driven Approaches
Our research focuses on improving speech processing algorithms, such as
automatic speech recognition (ASR), speaker identification, and depression
detection, under challenging conditions such as limited data (for example,
children’s or clinical speech), mismatched conditions (for example, training on
read speech while recognizing conversational speech), and noisy speech, using a hybrid data-driven and knowledge-based approach. This approach requires
understanding of both machine learning approaches and of the human speech
production and perception systems. I will summarize in this talk our work on
children’s ASR using self-supervised models, detecting depression from speech
signals using novel speaker disentanglement techniques, and automating scoring of children’s reading tasks with both ASR and innovative NLP algorithms.