Sam Keene
YOU?
Author Swipe
View article: BeamSeek: Deep Learning-based DOA Estimation for Low-Complexity mmWave Phased Arrays
BeamSeek: Deep Learning-based DOA Estimation for Low-Complexity mmWave Phased Arrays Open
A novel approach combining agile beam switching with deep learning to enhance the speed and accuracy of Direction of Arrival (DOA) estimation for millimeter-wave (mmWave) phased array systems with low-complexity hardware implementations is…
View article: Generative Algorithms for Art and Architecture: A Collaborative Teaching Approach
Generative Algorithms for Art and Architecture: A Collaborative Teaching Approach Open
We will present a course that we have been offering for the past few years that engages art, architecture and engineering students and challenges them to collaborate using generative methods to produce creative work. Our work contributes t…
View article: An Effective Automated Algorithm to Isolate Patient Speech from Conversations with Clinicians
An Effective Automated Algorithm to Isolate Patient Speech from Conversations with Clinicians Open
A growing number of algorithms are being developed to automatically identify disorders or disease biomarkers from digitally recorded audio of patient speech. An important step in these analyses is to identify and isolate the patient’s spee…
View article: MIRTH: Metabolite Imputation via Rank-Transformation and Harmonization
MIRTH: Metabolite Imputation via Rank-Transformation and Harmonization Open
Out of the thousands of metabolites in a given specimen, most metabolomics experiments measure only hundreds, with poor overlap across experimental platforms. Here, we describe Metabolite Imputation via Rank-Transformation and Harmonizatio…
View article: Additional file 1 of MIRTH: Metabolite Imputation via Rank-Transformation and Harmonization
Additional file 1 of MIRTH: Metabolite Imputation via Rank-Transformation and Harmonization Open
Additional file 1: Supplementary Tables S1-7. Compiled supplementary tables S1-7 with titles enclosed.
View article: Additional file 3 of MIRTH: Metabolite Imputation via Rank-Transformation and Harmonization
Additional file 3 of MIRTH: Metabolite Imputation via Rank-Transformation and Harmonization Open
Additional file 3: Example MIRTH embedding matrices. Sample and feature embedding matrices (W, H) generated by applying MIRTH to 9-dataset aggregate with 30 embedding dimensions.
View article: Autoencoding Neural Networks as Musical Audio Synthesizers
Autoencoding Neural Networks as Musical Audio Synthesizers Open
A method for musical audio synthesis using autoencoding neural networks is proposed. The autoencoder is trained to compress and reconstruct magnitude short-time Fourier transform frames. The autoencoder produces a spectrogram by activating…
View article: Conditioning Autoencoder Latent Spaces for Real-Time Timbre Interpolation and Synthesis
Conditioning Autoencoder Latent Spaces for Real-Time Timbre Interpolation and Synthesis Open
We compare standard autoencoder topologies' performances for timbre generation. We demonstrate how different activation functions used in the autoencoder's bottleneck distributes a training corpus's embedding. We show that the choice of si…
View article: A Fully Convolutional Neural Network Approach to End-to-End Speech Enhancement
A Fully Convolutional Neural Network Approach to End-to-End Speech Enhancement Open
This paper will describe a novel approach to the cocktail party problem that relies on a fully convolutional neural network (FCN) architecture. The FCN takes noisy audio data as input and performs nonlinear, filtering operations to produce…