Search results

This presentation demonstrates a practical, end-to-end workflow for constructing realistic acoustic datasets using state-of-the-art simulation software, with a focus on replacing traditional shoebox approximations with physically accurate environments for speech and audio AI model training.

This presentation explores the transformation of true wireless earbuds from basic audio playback devices into multimodal, health-monitoring wearable platforms. Driven by the unique anatomical advantages of the ear—including stable skin contact, rich local vasculature, and reduced motion interference from the musculoskeletal system—in-ear sensing is positioned to mirror the evolution of smartwatches toward continuous biometric tracking.

Dolby Atmos FlexConnect (DAFC) is an adaptive home?audio solution designed to operate with an arbitrary number of loudspeakers placed freely throughout a room, without requiring predefined layouts. The system begins by performing acoustic mapping to localize each speaker, estimating its position, distance, and orientation. These estimations feed into a flexible rendering framework that renders the Dolby Atmos or multichannel soundtrack in real time to maintain a faithful and stable soundstage across highly asymmetric speaker configurations.

This presentation addresses the continuous integration and evolution of artificial intelligence in signal processing within consumer electronics. Presented by Dr.

Sparse approximation is a well-established theory, with a profound impact on the fields of signal and image processing. In this talk, we start by presenting this model and its features, and then turn to describe two special cases from it: 1) the convolutional sparse coding (CSC) and 2) its multi-layered version (ML-CSC). Amazingly, as we will carefully show, ML-CSC provides a solid theoretical foundation to deep-learning architectures.

A person’s speech is strongly conditioned by his own articulators and the language(s) he speaks, hence, rendering speech in an interspeaker or interlanguage manner from a source speaker’s speech data collected in his native language that is both academically challenging and technology/application desirable. The quality of the rendered speech is assessed in three dimensions: naturalness, intelligibility, and similarity to the source speaker. Usually, the three criteria cannot be all met when rendering is done in both cross-speaker and cross-language ways.

We will present our paper 'Differentiable Uncalibrated Imaging' where we propose a differentiable imaging framework to address uncertainty in measurement coordinates, such as sensor locations and projection angles. Our approach leverages implicit neural networks and spline-based neural fields to perform joint calibration and image reconstruction. This makes it possible, for example, to use pre-trained neural networks to configurations different from those for which they were originally trained. We will begin by presenting the general optimization framework underlying our method.

Lensless cameras possess significant advantages when compared to lens-based ones. By replacing the lens with a coded aperture, the camera device can be made lighter, smaller, amenable towards compressive sensing techniques, and can be designed with less physical constraints. We leverage this freedom of design to propose a novel optimized coded mask that effectively extends the depth of field of a lensless camera. In this webinar, we will start by introducing the concept of lensless imaging along with the reason as to why general lensless cameras have a limited depth of field.