University Homepage School Homepage School Contacts School Search

ASR 2024-25  |  Lectures  |  Labs  |  Coursework  |  Piazza

Automatic Speech Recognition (ASR) 2024-25: Labs

There will be weekly lab sessions using OpenFst and Python to build speech recognition systems. Lab sessions start in Week 3, and will run in Appleton Tower in room 5.05. You need only attend one lab session per week – you should already have been allocated one of the sessions.

• Lab group 01: Tuesdays 10.00
• Lab group 05: Tuesdays 11.10
• Lab group 02: Wednesdays 10.00
• Lab group 03: Wednesdays 11.10

Labs are accessed via a GitHub repository with new exercises added at the start of each week. Setup instructions are contained in the README file.

Labs 1-5 are worth 10% of the total mark for the course. You will work in pairs, with each pair submitting one set of solutions per lab. Each lab submission is due at 9am on the Monday morning of the second week following the lab – details on Learn. Following the first five labs, there are three optional labs teaching the Kaldi ASR toolkit.

An introduction to the labs is presented in Lecture 5 (slides).

An introduction to the labs will be presented in Lecture 5.

Compulsory labs (via GitHub)

• Lab 1 (Week 3): Building HMMs with OpenFst
• Lab 2 (Week 4): Computing HMM probabilities
• Lab 3 (Week 5): Viterbi decoding (1)
• Lab 4 (Week 6): Viterbi decoding (2)
• Lab 5 (Week 7): WFST operations

Optional Kaldi labs (PDFs)

• Lab 6 (Week 8): Kaldi: data preparation and feature extraction
• Lab 7 (Week 9): Kaldi: training monophone models
• Lab 8 (Week 10): Kaldi: training triphone models

---

Copyright (c) University of Edinburgh 2015-2025
The ASR course material is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Public License
licence.txt
This page maintained by Peter Bell.
Last updated: 2025/03/10 22:02:05UTC