Tamil TTS Using Tacotron 2

The main goal of this project is building the Tamil TTS using Tacotron 2.

Pre-requisites

1. NVIDIA GPU + CUDA cuDNN

Setup

1. Download and extract the [Common Voice Dataset] https://commonvoice.mozilla.org/ta/datasets
2. Clone this repo: git clone https://github.com/vglug/Tamil-TTS-Using-tacotron2.git
3. Create a input text file with the path of the audio files and the sentence of the audio
4. Install PyTorch 1.0
5. Install Apex
6. Install python requirements or build docker image
   • Install python requirements: pip install -r requirements.txt

Training

1. python train.py --output_directory=outdir --log_directory=logdir
2. (OPTIONAL) tensorboard --logdir=outdir/logdir

Inference demo

1. Give your tained model
2. Download WaveGlow model
3. jupyter notebook --ip=127.0.0.1 --port=31337
4. Load inference_for_tamil_language.ipynb
5. Execute the steps one by one in the final step you will the audio of the given text

Related repos

WaveGlow Faster than real time Flow-based Generative Network for Speech Synthesis

nv-wavenet Faster than real time WaveNet.

Acknowledgements

This implementation uses code from the following repos: Keith Ito, Prem Seetharaman as described in our code.

We are inspired by Ryuchi Yamamoto's Tacotron PyTorch implementation.

We are thankful to the Tacotron 2 paper authors, specially Jonathan Shen, Yuxuan Wang and Zongheng Yang.