PaddleSpeech/examples/ljspeech/voc5

HiFiGAN with the LJSpeech-1.1

This example contains code used to train a HiFiGAN model with LJSpeech-1.1.

Dataset

Download and Extract

Download LJSpeech-1.1 from the official website.

Get MFA Result and Extract

We use MFA results to cut the silence in the edge of audio. You can download from here ljspeech_alignment.tar.gz, or train your MFA model reference to mfa example of our repo.

Get Started

Assume the path to the dataset is ~/datasets/LJSpeech-1.1. Assume the path to the MFA result of LJSpeech-1.1 is ./ljspeech_alignment. Run the command below to

1. source path.
2. preprocess the dataset.
3. train the model.
4. synthesize wavs.
   • synthesize waveform from metadata.jsonl.

./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.

./run.sh --stage 0 --stop-stage 0

Data Preprocessing

./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.

dump
├── dev
│   ├── norm
│   └── raw
├── test
│   ├── norm
│   └── raw
└── train
    ├── norm
    ├── raw
    └── feats_stats.npy

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 the log magnitude of the mel spectrogram of each utterance, while the norm folder contains the normalized spectrogram. The statistics used to normalize the spectrogram 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 id and paths to the spectrogram of each utterance.

Model Training

./local/train.sh calls ${BIN_DIR}/train.py.

CUDA_VISIBLE_DEVICES=${gpus} ./local/train.sh ${conf_path} ${train_output_path}

Here's the complete help message.

usage: train.py [-h] [--config CONFIG] [--train-metadata TRAIN_METADATA]
                [--dev-metadata DEV_METADATA] [--output-dir OUTPUT_DIR]
                [--ngpu NGPU] [--batch-size BATCH_SIZE] [--max-iter MAX_ITER]
                [--run-benchmark RUN_BENCHMARK]
                [--profiler_options PROFILER_OPTIONS]

Train a ParallelWaveGAN model.

optional arguments:
  -h, --help            show this help message and exit
  --config CONFIG       config file to overwrite default config.
  --train-metadata TRAIN_METADATA
                        training data.
  --dev-metadata DEV_METADATA
                        dev data.
  --output-dir OUTPUT_DIR
                        output dir.
  --ngpu NGPU           if ngpu == 0, use cpu.

benchmark:
  arguments related to benchmark.

  --batch-size BATCH_SIZE
                        batch size.
  --max-iter MAX_ITER   train max steps.
  --run-benchmark RUN_BENCHMARK
                        runing benchmark or not, if True, use the --batch-size
                        and --max-iter.
  --profiler_options PROFILER_OPTIONS
                        The option of profiler, which should be in format
                        "key1=value1;key2=value2;key3=value3".

1. --config is a config file in yaml format to overwrite the default config, which can be found at conf/default.yaml.
2. --train-metadata and --dev-metadata should be the metadata file in the normalized subfolder of train and dev in the dump folder.
3. --output-dir is the directory to save the results of the experiment. Checkpoints are saved in checkpoints/ inside this directory.
4. --ngpu is the number of gpus to use, if ngpu == 0, use cpu.

Synthesizing

./local/synthesize.sh calls ${BIN_DIR}/../synthesize.py, which can synthesize waveform from metadata.jsonl.

CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh ${conf_path} ${train_output_path} ${ckpt_name}

usage: synthesize.py [-h] [--generator-type GENERATOR_TYPE] [--config CONFIG]
                     [--checkpoint CHECKPOINT] [--test-metadata TEST_METADATA]
                     [--output-dir OUTPUT_DIR] [--ngpu NGPU]

Synthesize with GANVocoder.

optional arguments:
  -h, --help            show this help message and exit
  --generator-type GENERATOR_TYPE
                        type of GANVocoder, should in {pwgan, mb_melgan,
                        style_melgan, } now
  --config CONFIG       GANVocoder config file.
  --checkpoint CHECKPOINT
                        snapshot to load.
  --test-metadata TEST_METADATA
                        dev data.
  --output-dir OUTPUT_DIR
                        output dir.
  --ngpu NGPU           if ngpu == 0, use cpu.

1. --config parallel wavegan config file. You should use the same config with which the model is trained.
2. --checkpoint is the checkpoint to load. Pick one of the checkpoints from checkpoints inside the training output directory.
3. --test-metadata is the metadata of the test dataset. Use the metadata.jsonl in the dev/norm subfolder from the processed directory.
4. --output-dir is the directory to save the synthesized audio files.
5. --ngpu is the number of gpus to use, if ngpu == 0, use cpu.

Pretrained Model

The pretrained model can be downloaded here:

• hifigan_ljspeech_ckpt_0.2.0.zip

Model	Step	eval/generator_loss	eval/mel_loss	eval/feature_matching_loss
default	1(gpu) x 2500000	24.492	0.115	7.227

HiFiGAN checkpoint contains files listed below.

hifigan_ljspeech_ckpt_0.2.0
├── default.yaml                  # default config used to train hifigan
├── feats_stats.npy               # statistics used to normalize spectrogram when training hifigan
└── snapshot_iter_2500000.pdz     # generator parameters of hifigan

Acknowledgement

We adapted some code from https://github.com/kan-bayashi/ParallelWaveGAN.