A simple but functional digit recogniser built from scratch: record and label data, train HMMs, create a language model, and recognise the test data. Extend to other speakers & digit sequences.
Read all the instructions right the way through before you start: from the introduction right through to “digit sequences”.
Remember that the milestones will help you stay on track with this assignment.
Getting Started
Here are the files you need to get started and tips for how/where to do the assignment
Speaker-dependent system
The main task here is gathering the data. After that, just run the provided scripts.
Digit sequences (optional)
This is not as hard as you might think. If you attempt this, restrict your experiments to a speaker-dependent system using either your own data (if you collected some) or that from a single existing speaker in the database.
Writing up
You're going to write up a lab report, that describes how HMM-based ASR works, and reports your experiments.
