LinkedIn

I Trained a Neural Network Directly on a Microcontroller — No Cloud, No Server, Just the Chip

From a research paper to a working on-device TinyML system in a week. I am an embedded AI engineer. On most days, I find myself writing firmware, working on the design of printed circuit boards, and ...
IEEE

Real-Time Control of a Low-Cost Robotic Arm Using EMG Signal Classification by AI-Based Machine Learning on Raspberry Pi

Abstract: There are more than 804.5 million elderly people who need support for the normal functioning of their daily routine, and one in every three people suffers from some form of neurological ...
Scientific Research Publishing

Cybersecurity and Forensic Audio Analysis: Deepfake Detection Based on MFCC, Audio-Text Disconsistency, and Prosodic Features ()

1 Department of Computer and Instructional Technologies Education, Gazi Faculty of Education, Gazi University, Ankara, Türkiye. 2 Department of Forensic Informatics, Institute of Informatics, Gazi ...
Frontiers

Noninvasive prescreening of pediatric adenoid hypertrophy using quantified MFCC statistics and clinical features: development and external validation

In a dual-center cross-sectional study (N = 202), Center 1 (Capital Center for Children’s Health, Capital Medical University, n = 161) served as the development cohort and Center 2 (College of ...
IEEE

Classroom Acoustic Event Detection System

Abstract: Classroom environments often struggle with maintaining optimal acoustic conditions for effective learning. Traditional monitoring methods are manual, time-consuming, and unreliable. This ...
Analytics Insight

10 Innovative Keras Project Ideas with Source Code for 2025

Keras remains one of the most intuitive and developer-friendly frameworks for building deep learning models, making it perfect for learners and professionals alike. These 2025 project ideas cover ...
GitHub

HTK features in Python

HTK is a respected toolkit used mainly by the speech community to perform research in speech recognition. Although quite old, many newer systems emulate the same feature extraction pipeline as used in ...
GitHub

Continuous speech recognition with ESP32 for numbers (0--9) using a 1D CNN / MFCC

With this project I want to try out if its possible to detect the spoken numbers "zero" to "nine" by using an ESP32 processor and MEMS microphone and ML 1D CNN network While there are a lot examples ...
Nature

Novel transfer learning based acoustic feature engineering for scene fake audio detection

Audio forensics plays a major role in the investigation and analysis of audio recordings for legal and security purposes. The advent of audio fake attacks using speech combined with scene-manipulated ...
C&EN

Protein–Ligand Interaction Energies from Quantum-Chemical Fragmentation Methods: Upgrading the MFCC-Scheme with Many-Body Contributions

You are free to share (copy and redistribute) this article in any medium or format and to adapt (remix, transform, and build upon) the material for any purpose, even commercially within the parameters ...
Frontiers

Enhanced classification and severity prediction of major depressive disorder using acoustic features and machine learning

Previous studies have classified major depression and healthy control groups based on vocal acoustic features, but the classification accuracy needs to be improved. Therefore, this study utilized deep ...