Skip to content

Latest commit

 

History

History

pytorch-dc-tts

Efficiently Trainable Text-to-Speech System Based on Deep Convolutional Networks with Guided Attention

Input

A sentence which is defined as SENTENCE in pytorch-dc-tts.py.

Output

The Voice file is output as .wav which path is defined as SAVE_WAV_PATH in pytorch-dc-tts.py.

Usage

Automatically downloads the onnx and prototxt files on the first run. It is necessary to be connected to the Internet while downloading.

SENTENCE is defined in the pytorch-dc-tts.py. ex. SENTENCE = 'The boy was there when the sun rose.'

For the sample sentence,

python3 pytorch-dc-tts.py 

If you want to specify the input sentence, put the wav path after the --input option. You can use --savepath option to change the name of the output file to save.

python3 pytorch-dc-tts.py  --input SENTENCE --savepath SAVE_WAV_PATH

Reference

Efficiently Trainable Text-to-Speech System Based on Deep Convolutional Networks with Guided Attention

Framework

PyTorch

Model Format

ONNX opset = 10

Netron