To download vocoder, add

"vocoder": {
        "path": "<path_to_be_placed>"
},

to config.json It will be waveglow v2 trained on ljspeech.

I used pretrained MFA. You might download it here or just add

"preprocessing": {
   ...
   "g2p": true,
   ...
}

to config. It will be placed in the same directory, as your provided dataset.

• A defibrillator is a device that gives a high energy electric shock to the heart of someone who is in cardiac arrest
• Massachusetts Institute of Technology may be best known for its math, science and engineering education
• Wasserstein distance or Kantorovich Rubinstein metric is a distance function defined between probability distributions on a given metric space

These examples of synthezed audio are provided in directory test_data

You may see the report if you follow this link

Firstly, install needed requirements for running model

pip install -r ./requirements.txt

Use bash script to download trained model

cd ./default_test_model
./download.sh

It will be placed to ./default_test_model/checkpoint.pth

If you have some issues using bash utilities, you may download model directly from google drive

python test.py \
   -c default_test_config.json \
   -r default_test_model/checkpoint.pth \
   -t test_data \
   -o test_result.json

This repository is based on a heavily modified fork of pytorch-template repository.

You can use this project with docker. Quick start:

docker build -t my_src_image .
docker run \
   --gpus '"device=0"' \
   -it --rm \
   -v /path/to/local/storage/dir:/repos/asr_project_template/data/datasets \
   -e WANDB_API_KEY=<your_wandb_api_key> \
	my_src_image python -m unittest

Notes:

• -v /out/of/container/path:/inside/container/path -- bind mount a path, so you wouldn't have to download datasets at the start of every docker run.
• -e WANDB_API_KEY=<your_wandb_api_key> -- set envvar for wandb (if you want to use it). You can find your API key here: https://wandb.ai/authorize