Speech Artificial Intelligence

Module overview

Speech is humanity's most natural interface. Speech is all around us, but have you ever wondered what a voice really is, and what it takes to build one? In this module, you will develop expertise across the full speech AI pipeline: from speech signal fundamentals and data preparation to synthesis architectures that generate remarkably human voices. You will explore the theory and AI architectures that underlie state of the art speech technologies. Building on your knowledge of machine learning, this module dives deeper into open and unsolved problems in the field, driven by cutting edge research. You will learn why speech processing is challenging and you will use creative problem solving to explore both the limitations and promises of speech AI.

Linked modules

prerequisites: COMP6246 Machine Learning Technologies

Aims and Objectives

Learning Outcomes

Subject Specific Intellectual and Research Skills

Having successfully completed this module you will be able to:

Propose improvements to existing state of the art speech AI algorithms.
Engage critically with recent speech technology literature, including key advances and open problems.
Demonstrate familiarity with techniques that are used to conduct and design experiments.
Describe how to algorithmically model speech sounds and speaker attributes.
Critique, compare, and interpret the performance of speech AI algorithms.
Explain how speech AI algorithms can solve a problem or test a hypothesis.

Knowledge and Understanding

Having successfully completed this module, you will be able to demonstrate knowledge and understanding of:

Key challenges and limitations underpinning speech technology performance across diverse speakers and languages.
The fundamental building blocks of human voice.
The algorithmic and theoretical basis for modelling speech and speakers.

Learning and Teaching

Teaching and learning methods

Lectures, labs, self-guided study

Study time
Type	Hours
Specialist Laboratory	8
Preparation for scheduled sessions	12
Completion of assessment task	50
Wider reading or practice	16
Revision	40
Lecture	24
Total study time	150

Resources & Reading list

Internet Resources

Speech Synthesis and Recognition.

Speech and Language Processing.

Text to Speech Synthesis.

Assessment

Summative

This is how we’ll formally assess what you have learned in this module.

Breakdown
Method	Percentage contribution
Coursework	40%
Exam	60%

Referral

This is how we’ll assess you if you don’t meet the criteria to pass this module.

Breakdown
Method	Percentage contribution
Exam	100%

Repeat

An internal repeat is where you take all of your modules again, including any you passed. An external repeat is where you only re-take the modules you failed.

Breakdown
Method	Percentage contribution
Exam	100%

Repeat Information

Repeat type: Internal & External