Academy for Continuing Education (ACE)

Speech recognition is the process by which a computer, or another type of machine, identifies spoken words and transcribes them into text.

In this short course, participants will learn the basics of digital signal processing, understanding syntactic and semantic analyses, word embeddings, speech recognition processes, and various machine learning models to train automatic speech recognition. You will learn to develop simple speech recognition and synthesis applications using Python. Participants will also get the chance to build a customised speech recognition engine from scratch using the Kaldi speech recognition toolkit.

Who should attend?

Software engineers, Information Technology professionals, and Artificial Intelligence enthusiasts.

Learning Objectives

By the end of this course, participants will be able to:

Speech Recognition Basics

• Understand the fundamentals of digital signal processing (DSP)
• Explain the overview of natural language processing (NLP) technologies
• Describe the applications of NLP
• Identify speech recognition processes
• Distinguish the types of automatic speech recognition (ASR)
• Understand Gaussian Mixture Model (GMM), Hidden Markov Model (HMM), and Recurrent Neural Network (RNN) for use as a classifier

Using Speech Recognition via pre-trained Google Application Programming Interface

• Develop and test a speech transcription (speech-to-text) application using Python
• Develop and test a speech synthesis (text-to-speech) application using Python

Building Your Customised Speech Recognition Engine

• Compile and build your Kaldi ASR on Ubuntu
• Train, test and validate the accuracy of audio data set using Kaldi

Entry Requirement

Diploma/Degree from any Information and Communication Technology, and Engineering courses.

Certification

Participants will be awarded a certificate of completion upon meeting the 75% course attendance requirement.

For courses with assessment component, participants will be awarded the certification of completion upon passing the assessment. Otherwise, a certification of attendance will be issued instead upon meeting the 75% course attendance requirement.