microteaching

Basics of Speech Recognition

-Aims
Speech recognition has gained considerable research focus in recent years. Technologies, such as Siri, have begun proliferating the consumer markets but there still remain plenty of challenges in this field. The aim of this course is to introduce the topic of speech recognition and give an intuitive understanding of the concepts behind modern speech recognition systems. After this course, you would be also be aware of the challenges faced in this field.

-Learning Outcomes
.Identify the different disciplines behind building a working speech recognition system in computers
.Understand how Phonemes and Hidden Markov Models are used to model languages
.Understand how these language models are used to detect patterns in speech recognition using search algorithms
.Realise the challenges behind achieving effective speech recognition, based on accuracy and responsiveness metrics

-Learning Activities
.You will get a chance to meddle with a speech recognition application on an Android Tablet
.You will get a firsthand experience of observing the inaccuracies of current open source speech recognition software
.You will also see how different language model sizes can affect your speech recognition software performance in real-time

-Resources
.PC with PDF support (presentation slides in PDF format)
.Tablet with application installed (provided by lecturer)