Ivan Kunyankin

# Description Functions used inside f1_score function use `beta` as a parameter. However, f1_score function itself doesn't give the opportunity to change the balance between precision and recall. It could...

Hi, thanks for the great work! Could you please provide a step-by-step guidance of how to run the model on new data in each mode (there are several of them...

Things that are different compared to the article

1

There are a number of differences compared to the source: * original article: https://arxiv.org/pdf/1910.10261.pdf Differences: 1. The default training script uses LibriTTS dataset instead of LibriSpeech. 2. The model implementation...

By default the repo can train either version of the model on the clean part of LibriTTS. It has lists of audio files with corresponding labels and the file containing...

A short description of hyperparameters from config.yml ### Training parameters 1. ```max_length``` - maximum file length in seconds. Longer files are omitted. Used to avoid OOM (out-of-memory) error. Also reduces...