|
2 years ago | |
---|---|---|
.. | ||
conf | 2 years ago | |
local | 2 years ago | |
README.md | 2 years ago | |
path.sh | 2 years ago | |
run.sh | 2 years ago |
README.md
HiFiGAN with AISHELL-3
This example contains code used to train a HiFiGAN model with 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.
Dataset
Download and Extract
Download AISHELL-3.
wget https://www.openslr.org/resources/93/data_aishell3.tgz
Extract AISHELL-3.
mkdir data_aishell3
tar zxvf data_aishell3.tgz -C data_aishell3
Get MFA Result and Extract
We use MFA2.x to get durations for aishell3_fastspeech2. You can download from here aishell3_alignment_tone.tar.gz, or train your MFA model reference to mfa example (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
- source path.
- preprocess the dataset.
- train the model.
- synthesize wavs.
- synthesize waveform from
metadata.jsonl
.
- synthesize waveform from
./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, run the following command will only preprocess the dataset.
./run.sh --stage 0 --stop-stage 0
Data Preprocessing
./local/preprocess.sh ${conf_path}
When it is done. A dump
folder is created in the current directory. The structure of the dump folder is listed below.
dump
├── dev
│ ├── norm
│ └── raw
├── test
│ ├── norm
│ └── raw
└── train
├── norm
├── raw
└── feats_stats.npy
The dataset is split into 3 parts, namely train
, dev
, and test
, each of which contains a norm
and raw
subfolder. The raw
folder contains the log magnitude of the mel spectrogram of each utterance, while the norm folder contains the normalized spectrogram. The statistics used to normalize the spectrogram are computed from the training set, which is located in dump/train/feats_stats.npy
.
Also, there is a metadata.jsonl
in each subfolder. It is a table-like file that contains id and paths to the spectrogram of each utterance.
Model Training
CUDA_VISIBLE_DEVICES=${gpus} ./local/train.sh ${conf_path} ${train_output_path}
./local/train.sh
calls ${BIN_DIR}/train.py
.
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".
--config
is a config file in yaml format to overwrite the default config, which can be found atconf/default.yaml
.--train-metadata
and--dev-metadata
should be the metadata file in the normalized subfolder oftrain
anddev
in thedump
folder.--output-dir
is the directory to save the results of the experiment. Checkpoints are saved incheckpoints/
inside this directory.--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.
--config
config file. You should use the same config with which the model is trained.--checkpoint
is the checkpoint to load. Pick one of the checkpoints fromcheckpoints
inside the training output directory.--test-metadata
is the metadata of the test dataset. Use themetadata.jsonl
in thedev/norm
subfolder from the processed directory.--output-dir
is the directory to save the synthesized audio files.--ngpu
is the number of gpus to use, if ngpu == 0, use cpu.
Pretrained Models
The pretrained model can be downloaded here:
Model | Step | eval/generator_loss | eval/mel_loss | eval/feature_matching_loss |
---|---|---|---|---|
default | 1(gpu) x 2500000 | 24.060 | 0.1068 | 7.499 |
HiFiGAN checkpoint contains files listed below.
hifigan_aishell3_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.