collaborators

5 papers

eess.AS2025

Switchboard-Affect: Emotion Perception Labels from Conversational Speech

Amrit Romana, Jaya Narain, Tien Dung Tran +4

Understanding the nuances of speech emotion dataset curation and labeling is essential for assessing speech emotion recognition (SER) model potential in real-world applications. Mo…

cs.LG2025

Affect Models Have Weak Generalizability to Atypical Speech

Jaya Narain, Amrit Romana, Vikramjit Mitra +2

Speech and voice conditions can alter the acoustic properties of speech, which could impact the performance of paralinguistic models for affect for people with atypical speech. We…

cs.SD2025

Foundation Model Hidden Representations for Heart Rate Estimation from Auscultation

Jingping Nie, Dung T. Tran, Karan Thakkar +5

Auscultation, particularly heart sound, is a non-invasive technique that provides essential vital sign information. Recently, self-supervised acoustic representation foundation mod…

cs.SD2025

Voice Quality Dimensions as Interpretable Primitives for Speaking Style for Atypical Speech and Affect

Jaya Narain, Vasudha Kowtha, Colin Lea +8

Perceptual voice quality dimensions describe key characteristics of atypical speech and other speech modulations. Here we develop and evaluate voice quality models for seven voice…

cs.SD2025

Modeling speech emotion with label variance and analyzing performance across speakers and unseen acoustic conditions

Vikramjit Mitra, Amrit Romana, Dung T. Tran +1

Spontaneous speech emotion data usually contain perceptual grades where graders assign emotion score after listening to the speech files. Such perceptual grades introduce uncertain…