This is a checkpoint for the Tacotron 2 model that was trained in NeMo on LJspeech for 1200 epochs. It was trained with Apex/Amp optimization level O0, with 8 * 16GB V100, and with a batch size of 48 per GPU for a total batch size of 384.
It contains the checkpoints for the Tacotron 2 Neural Modules and the yaml config file:
Refer to documentation at https://github.com/NVIDIA/NeMo
Usage example: Put the checkpoints into the checkpoint dir, and run tts_infer.py (from NeMo's TTS examples).
python tts_infer.py --model_config=$checkpoint_dir/tacotron2.yaml --eval_dataset=test.json --load_dir=$checkpoint_dir