CS224S Assignment 4: SpeechBrain ASR and Voice Cloning

Spring 2022

Please read this entire handout before beginning. We advise you to start early (training our model should take >1 hour) and to make use of the TAs by coming to office hours and asking questions! For collaboration and the late day policy, please refer to the home page.

About the Assignment

In this assignment you’ll walk through a example training and evaluating a SpeechBrain ASR system, as well as some exercises with a voice cloning toolkit. The goal of this assignment is to introduce you to the PyTorch-based SpeechBrain framework that may be useful for your projects, so we hope that it’s lighter than usual.

Submission Instructions

This assignment is due on 05/23/2022 by 11:59PM pacific (or at latest on 05/26/2022 with three late days). For both parts of this assignment, you will submit

  • Your filled-in/executed Colab Notebooks with all code/output, as PDF (combine both PDFs into the same file) to Assignment 4: PDF Submission on Gradescope. Please tag your question responses.
  • The source notebooks (zipped together, if needed) with all cells fully executed to Assignment 4: Code Submission on Gradescope.

All instructions and starter code are contained in the Google Colab notebooks. You will need to make a copy to your Google Drive before beginning. There are two notebooks required for this assignment:

  • SpeechBrain ASR notebook
  • Voice Cloning notebook

You can access the two starter notebooks in the Assignment 4 directory. Remember to make a copy before starting your work!