IEEE

CleanMel: Mel-Spectrogram Enhancement for Improving Both Speech Quality and ASR

Abstract: In this work, we propose CleanMel, a single-channel Mel-spectrogram denoising and dereverberation network for improving both speech quality and automatic speech recognition (ASR) performance ...
GitHub

Making Acoustic Side-Channel Attacks on Noisy Keyboards Viable with LLM-Assisted Spectrograms "Typo" Correction

This repo is the implementation of a research project aimed at enhancing Acoustic Side-Channel Attacks (ASCAs) using a novel combination of Vision Transformers (VTs) and Large Language Models (LLMs).
GitHub

songsee — FFT so pretty, your ears will be jealous.

--output Output path (default: input name + extension) --format jpg or png (default: jpg) --width Output width (default: 1920) --height Output height (default: 1080 ...
IEEE

A High Accuracy and Ultra-Low Energy Environmental Sound Recognition Processor with Progressive Spectrogram Processing and Adaptive Weight Clustering Based Online Learning

Abstract: In recent years, environmental sound recognition (ESR) technology has been proposed to detect sounds of interest in the environment [1] [2], which is useful in many applications such as ...