Kalender

Introduction and Speech Production

Here are the Slides for Lecture 01

The speech analysis program I talked about is called Wavesurfer (https://sourceforge.net/projects/wavesurfer/). If you are running a Debian based OS such as Ubuntu, you can install with:

sudo apt-get install libsnack-alsa wavesurfer

Reading Instructions:

Huang Acero and Hon:

• Chapter 1
• Chapter 2 (2.3 - 2.5 are optional)

Signal Processing and Feature Extraction

Here are the Slides for Lecture 02

Reading Instructions:

Huang Acero and Hon:

• Chapter 5 (review of signal processing)
• Chapter 6 (Speech signal representations, filterbanks and MFCCs)

Machine Learning Reminder

Here are the Slides for lecture 03

Reading instructions:

From Huang Acero and Hon

• Chapter 3 (fast review of probability theory and estimation theory)
• Chapter 4 (fast review of Pattern Recognition)

Modelling Sequences

Here are the Slides for Lecture 04

Modelling Sequences (continued)

The slides are the same as for the previous lecture.

Here are my notes on how to compute the recursion formulas for alpha, beta and gamma

Acoustic Modelling

Here are the Slides for lecture 06

Deep Learning and Acoustic Modelling

Here are the Slides for Lecture 07 (I will probably update the end of the presentation before next lecture).

Deep Learning (Continued)

Introduction to the Parallel Data Centre

Introduction to PDC Slides

NOTE: the myCAC in the slides (time allocation code) is edu16.DT2118 for this course

Adaptation and Noise Robustness

Here are the Slides

Language Models

Here are the Slides on Language Modelling

 Search Algorithms

Here are the Slides on Search Algorithms

Speaker Recognition

Here are the Slides on Speaker Recognition