candywal/text-to-speech

Project for CSC 466

text-to-speech

Project for CSC 466

notes

We use semi-supervised learning to represent speech. We teech speech input and then output a transcribed speech.

We are basing it on the wav2vec 2.0 paper.

Run using:

pip install -r requirements.txt
python3 driver.py

resources

https://github.com/candywal/text-to-speech.git https://arxiv.org/pdf/2006.11477.pdf https://www.youtube.com/watch?v=aUSXvoWfy3w