Course abstract - van ESCH

Course: Machines can do the work, can't they? Machine Learning for linguists 

Instructors: Daan van Esch (Google) and guests

Description: Machine Learning promises great workflow improvements for language workers, offering deeper insights into research data and new ways to answer research questions. But just what is Machine Learning (ML), and how can linguists use it? This course introduces a range of ML topics, from speech-to-text, neural machine translation, and text-to-speech. Each topic will begin with an introduction, followed by a practical session, and will be suitable for beginners as well as people who have computational experience. Please bring your own language data if you have some – data specifications will be provided.


1. Overview of Automatic Speech Recognition and Machine Learning for Linguistics

2. Text-to-speech systems

3. Speech-to-text systems using Kaldi, Persephone

4. Neural Machine Translation

Note: This course has two sessions per day, in the second and fourth slots (11am-12:30pm and 3:30-5pm)

  • Australian Government
  • The University of Queensland
  • Australian National University
  • The University of Melbourne
  • Western Sydney University