This example contains code used to train a [VITS](https://arxiv.org/abs/2106.06103) model with [AISHELL-3](http://www.aishelltech.com/aishell_3).
AISHELL-3 is a large-scale and high-fidelity multi-speaker Mandarin speech corpus that could be used to train multi-speaker Text-to-Speech (TTS) systems.
We use AISHELL-3 to train a multi-speaker VITS model here.
## Dataset
### Download and Extract
Download AISHELL-3 from it's [Official Website](http://www.aishelltech.com/aishell_3) and extract it to `~/datasets`. Then the dataset is in the directory `~/datasets/data_aishell3`.
### Get MFA Result and Extract
We use [MFA2.x](https://github.com/MontrealCorpusTools/Montreal-Forced-Aligner) to get phonemes for VITS, the durations of MFA are not needed here.
You can download from here [aishell3_alignment_tone.tar.gz](https://paddlespeech.bj.bcebos.com/MFA/AISHELL-3/with_tone/aishell3_alignment_tone.tar.gz), or train your MFA model reference to [mfa example](https://github.com/PaddlePaddle/PaddleSpeech/tree/develop/examples/other/mfa) (use MFA1.x now) of our repo.
## Get Started
Assume the path to the dataset is `~/datasets/data_aishell3`.
Assume the path to the MFA result of AISHELL-3 is `./aishell3_alignment_tone`.
Run the command below to
1.**source path**.
2. preprocess the dataset.
3. train the model.
4. synthesize wavs.
- synthesize waveform from `metadata.jsonl`.
- synthesize waveform from a text file.
```bash
./run.sh
```
You can choose a range of stages you want to run, or set `stage` equal to `stop-stage` to use only one stage, for example, running the following command will only preprocess the dataset.
```bash
./run.sh --stage 0 --stop-stage 0
```
### Data Preprocessing
```bash
./local/preprocess.sh ${conf_path}
```
When it is done. A `dump` folder is created in the current directory. The structure of the dump folder is listed below.
```text
dump
├── dev
│ ├── norm
│ └── raw
├── phone_id_map.txt
├── speaker_id_map.txt
├── test
│ ├── norm
│ └── raw
└── train
├── feats_stats.npy
├── norm
└── raw
```
The dataset is split into 3 parts, namely `train`, `dev`, and` test`, each of which contains a `norm` and `raw` subfolder. The raw folder contains wave and linear spectrogram of each utterance, while the norm folder contains normalized ones. The statistics used to normalize features are computed from the training set, which is located in `dump/train/feats_stats.npy`.
Also, there is a `metadata.jsonl` in each subfolder. It is a table-like file that contains phones, text_lengths, feats, feats_lengths, the path of linear spectrogram features, the path of raw waves, speaker, and the id of each utterance.
--spk_id SPK_ID spk id for multi speaker acoustic model
--lang LANG Choose model language. zh or en
--inference_dir INFERENCE_DIR
dir to save inference models
--ngpu NGPU if ngpu == 0, use cpu.
--text TEXT text to synthesize, a 'utt_id sentence' pair per line.
--output_dir OUTPUT_DIR
output dir.
```
1.`--config`, `--ckpt`, `--phones_dict` and `--speaker_dict` are arguments for acoustic model, which correspond to the 3 files in the VITS pretrained model.
2.`--lang` is the model language, which can be `zh` or `en`.
3.`--test_metadata` should be the metadata file in the normalized subfolder of `test` in the `dump` folder.
4.`--text` is the text file, which contains sentences to synthesize.
5.`--output_dir` is the directory to save synthesized audio files.
6.`--ngpu` is the number of gpus to use, if ngpu == 0, use cpu.
<!-- TODO display these after we trained the model -->