TALP Talk: Introduction to Deep Learning for Speech Processing

Introduction to Deep Learning for Speech Processing

date: Wednesday 13-09-2017 time: From 11:00 to 13:00

Room: Campus NORD UPC, Building D5, Classroom MERIT D5-010, Barcelona

As you know, Deep Learning has emerged as a revolution in areas such as computer vision and speech and language processing.

This session is an introduction to deep learning including the most used architectures for speech processing, which are nevertheless widely used in other areas.

It can be a good introduction for students who are now beginning their degree or master's work, although it is open to anyone interested, with some technical training.


1. Introduction to Deep Learning

2. Tutorial in Deep Learning: the multilayer perceptron classifier in "Keras" (logistic regression, deep neural networks, activation functions, optimizers, regularization)

3. Architectures: convolutional, recurrent, embeddings, sequence-by-sequence, generative adversarial networks

The course is based on some of the parts of the seminar
Deep Learning for Speech and Language,

This seminar will continue with several sessions of the pytorch development environment, taught by Santi Pascual.


Additional information