Speech Presence
Speech presence, encompassing the accurate representation and processing of speech signals in various contexts, is a multifaceted research area aiming to improve robustness and efficiency in applications like speech synthesis, recognition, and analysis. Current research focuses on developing models that handle challenges such as noisy or incomplete data, interference from other sources (e.g., background noise or multiple speakers), and the presence of biases or spurious correlations in training data, often employing deep learning architectures and causal inference methods. These advancements are crucial for creating more reliable and adaptable systems across diverse applications, ranging from improved human-computer interaction to more accurate medical diagnosis and environmental monitoring.
Papers
MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion
Junyi Zhang, Charles Herrmann, Junhwa Hur, Varun Jampani, Trevor Darrell, Forrester Cole, Deqing Sun, Ming-Hsuan Yang
In-context Learning in Presence of Spurious Correlations
Hrayr Harutyunyan, Rafayel Darbinyan, Samvel Karapetyan, Hrant Khachatrian