https://github.com/kozistr/tacotron-tensorflow

A TensorFlow implementation of Google's Tacotron speech synthesis
https://github.com/kozistr/tacotron-tensorflow

tacotron tensorflow tts

Last synced: about 2 months ago
JSON representation

A TensorFlow implementation of Google's Tacotron speech synthesis

• Host: GitHub
• URL: https://github.com/kozistr/tacotron-tensorflow
• Owner: kozistr
• License: mit
• Created: 2018-12-24T07:08:31.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2019-01-01T15:32:02.000Z (over 7 years ago)
• Last Synced: 2025-10-24T22:34:22.395Z (8 months ago)
• Topics: tacotron, tensorflow, tts
• Language: Python
• Homepage:
• Size: 866 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# tacotron-tensorflow

A TensorFlow implementation of DeepMind's Tacotron. A deep neural network architectures described in many papers.

Especially for English, Korean.

highly inspired by [here](https://github.com/Rayhane-mamah/Tacotron-2)

[![Total alerts](https://img.shields.io/lgtm/alerts/g/kozistr/tacotron-tensorflow.svg?logo=lgtm&logoWidth=18)](https://lgtm.com/projects/g/kozistr/tacotron-tensorflow/alerts/)

[![Language grade: Python](https://img.shields.io/lgtm/grade/python/g/kozistr/tacotron-tensorflow.svg?logo=lgtm&logoWidth=18)](https://lgtm.com/projects/g/kozistr/tacotron-tensorflow/context:python)

## Requirements

* Python 3.x (preferred)

* Tensorflow 1.x

* matplotlib

* librosa

* numpy

* tqdm

## Usage

### 0. Download Dataset

* [IJSpeech 1.1](https://keithito.com/LJ-Speech-Dataset/)

### 0. Install Pre-Requisites

``` python -m pip install -r requirements.txt ```

### 1. Adjust Configuration

``` edit config.py ```

### 2. Train!

``` python train.py ```

## DataSet

|          DataSet          |     Samples    |          Size                 |

| :-----------------------: | :------------: | :---------------------------: |

|       IJSpeech-1.1        |      13100     |   about 30GB is needed        |

## Source Tree

```

│

├── assets

│    └── images       (readme images)

├── datasets

│    ├── ljspeech.py  (LJSpeech 1.1 DataSet)

│    └── ...

├── model

│    └── log data     (readme images)

├── config.py         (whole configuration)

├── dataloader.py     (data loading stuff)

├── model.py          (lots of TTS models)

├── modules.py        (lots of modules frequently used at model)

├── synthesize.py     (inference)

├── train.py          (model training)

├── utils.py          (useful utils)

└── tfutils.py        (useful TF utils)

```

## Model Architecture

### Tacotron 1

![architecture](./assets/tacotron-1.png)

### Tacotron 2

![architecture](./assets/tacotron-2.png)

### DeepVoice v2

soon!

### DeepVoice v3

![architecture](./assets/deep_voice_3.png)

## Author

HyeongChan Kim / [@kozistr](http://kozistr.tech)