You can generate datasets for 3 public Korean figures including:
1. [Sohn Suk-hee](https://en.wikipedia.org/wiki/Sohn_Suk-hee): anchor and president of JTBC
2. [Park Geun-hye](https://en.wikipedia.org/wiki/Park_Geun-hye): a former President of South Korea
3. [Moon Jae-in](https://en.wikipedia.org/wiki/Moon_Jae-in): the current President of South Korea
Each dataset can be generated with following scripts:
./scripts/prepare_son.sh # Sohn Suk-hee
./scripts/prepare_park.sh # Park Geun-hye
./scripts/prepare_moon.sh # Moon Jae-in
Each script execute below commands. (explain with `son` dataset)
0. To automate an alignment between sounds and texts, prepare `GOOGLE_APPLICATION_CREDENTIALS` to use [Google Speech Recognition API](https://cloud.google.com/speech/). To get credentials, read [this](https://developers.google.com/identity/protocols/application-default-credentials).
3. By using [Google Speech Recognition API](https://cloud.google.com/speech/), we predict sentences for all segmented audios. (this is optional for `moon` and `park` because they already have `recognition.json`)
Because the automatic generation is extremely naive, the dataset is noisy. However, if you have enough datasets (20+ hours with random initialization or 5+ hours with pretrained model initialization), you can expect an acceptable quality of audio synthesis.
### 2-3. Download pre-trained models
You can download a pre-trained models or generate audio. Available models are:
1. Single speaker model for [Sohn Suk-hee](https://en.wikipedia.org/wiki/Sohn_Suk-hee).
python3 download.py son
2. Single speaker model for [Park Geun-hye](https://en.wikipedia.org/wiki/Park_Geun-hye).
python3 download.py park
After you donwload pre-trained models, you can generate voices as follows:
If you don't have good and enough (10+ hours) dataset, it would be better to use `--initialize_path` to use a well-trained model as initial parameters.
This is not an official [DEVSISTERS](http://devsisters.com/) product. This project is not responsible for misuse or for any damage that you may cause. You agree that you use this software at your own risk.