Are you a signal processing engineer or scientist looking to integrate AI capabilities in your project?
In this session you will learn the basics of deep learning for signals by walking through a detailed example of Acoustics – speech classification, entirely based on MATLAB code. We will cover creating and accessing labelled data, using time-frequency transformations, extracting features, designing and training deep neural network architectures, and testing prototypes, mentioning about speeding up computations with GPU. We will also discuss the interoperability with other popular deep learning tools, including exploiting available pre-trained networks.
- Acquiring, segmenting and labelling signals
- Extracting standard features using 2D time-frequency representations
- Designing and analysing deep networks and exchanging models with other popular frameworks (e.g. via ONNX)
- Accelerating computations using GPUs and prototyping trained models on real-world signals
|10 min||Introduction to Deep Learning|
|20 min||Walkthrough example on speech recognition|
|10 min||Real-world acoustic applications|
Who Should Attend?
MATLAB users who have at least a basic working knowledge of the MATLAB Environment will benefit from this session. Advanced users will learn about newer capabilities introduced in recent releases of MATLAB.
About the presenter
Dr. Martina Sciola joined MathWorks as a Technical Specialist Engineer for Education in 2018 after her PhD studies at the University of Sheffield.
With a background on Biomedical Engineering, Martina’s PhD focused on personalised cardiovascular modelling. She also had the opportunity to work as research associate for diagnosis of Pulmonary Hypertension and as research coordinator for an EU project to establish a new centre of excellence of medicine at the University of Sheffield. Her interests are Healthcare, Clinical applications, In silico medicine and Biomedical Engineering.
How to join this webinar
Follow this link to register for the seminar: www.mathworks.com/IntroToDeepLearningForSignals
Talks to follow
After this webinar, PhD student, Jamie Scanlan, will present two talks on his work at the University of Salford and on Independent Component Analysis. The slides are available here.