PaddleSpeech/examples/csmsc/tts3

FastSpeech2 with CSMSC

This example contains code used to train a Fastspeech2 model with Chinese Standard Mandarin Speech Copus.

Dataset

Download and Extract

Download CSMSC from it's Official Website.

Get MFA Result and Extract

We use MFA to get durations for fastspeech2. You can download from here baker_alignment_tone.tar.gz, or train your own MFA model reference to mfa example of our repo.

Get Started

Assume the path to the dataset is ~/datasets/BZNSYP. Assume the path to the MFA result of CSMSC is ./baker_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 text file.
5. inference using static model.

./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
├── phone_id_map.txt
├── speaker_id_map.txt
├── test
│   ├── norm
│   └── raw
└── train
    ├── energy_stats.npy
    ├── norm
    ├── pitch_stats.npy
    ├── raw
    └── speech_stats.npy

The dataset is split into 3 parts, namely train, dev and test, each of which contains a norm and raw sub folder. The raw folder contains speech、pitch and energy features of each utterances, 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/*_stats.npy.

Also there is a metadata.jsonl in each subfolder. It is a table-like file which contains phones, text_lengths, speech_lengths, durations, path of speech features, path of pitch features, path of energy features, speaker and id 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] [--verbose VERBOSE] [--phones-dict PHONES_DICT]
                [--speaker-dict SPEAKER_DICT]

Train a FastSpeech2 model.

optional arguments:
  -h, --help            show this help message and exit
  --config CONFIG       fastspeech2 config file.
  --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.
  --verbose VERBOSE     verbose.
  --phones-dict PHONES_DICT
                        phone vocabulary file.
  --speaker-dict SPEAKER_DICT
                        speaker id map file for multiple speaker model.

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 save in checkpoints/ inside this directory.
4. --ngpu is the number of gpus to use, if ngpu == 0, use cpu.
5. --phones-dict is the path of the phone vocabulary file.

Synthesizing

We use parallel wavegan as the neural vocoder. Download pretrained parallel wavegan model from pwg_baker_ckpt_0.4.zip and unzip it.

unzip pwg_baker_ckpt_0.4.zip

Parallel WaveGAN checkpoint contains files listed below.

pwg_baker_ckpt_0.4
├── pwg_default.yaml               # default config used to train parallel wavegan
├── pwg_snapshot_iter_400000.pdz   # model parameters of parallel wavegan
└── pwg_stats.npy                  # statistics used to normalize spectrogram when training parallel wavegan

./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] [--fastspeech2-config FASTSPEECH2_CONFIG]
                     [--fastspeech2-checkpoint FASTSPEECH2_CHECKPOINT]
                     [--fastspeech2-stat FASTSPEECH2_STAT]
                     [--pwg-config PWG_CONFIG]
                     [--pwg-checkpoint PWG_CHECKPOINT] [--pwg-stat PWG_STAT]
                     [--phones-dict PHONES_DICT] [--speaker-dict SPEAKER_DICT]
                     [--test-metadata TEST_METADATA] [--output-dir OUTPUT_DIR]
                     [--ngpu NGPU] [--verbose VERBOSE]

Synthesize with fastspeech2 & parallel wavegan.

optional arguments:
  -h, --help            show this help message and exit
  --fastspeech2-config FASTSPEECH2_CONFIG
                        fastspeech2 config file.
  --fastspeech2-checkpoint FASTSPEECH2_CHECKPOINT
                        fastspeech2 checkpoint to load.
  --fastspeech2-stat FASTSPEECH2_STAT
                        mean and standard deviation used to normalize
                        spectrogram when training fastspeech2.
  --pwg-config PWG_CONFIG
                        parallel wavegan config file.
  --pwg-checkpoint PWG_CHECKPOINT
                        parallel wavegan generator parameters to load.
  --pwg-stat PWG_STAT   mean and standard deviation used to normalize
                        spectrogram when training parallel wavegan.
  --phones-dict PHONES_DICT
                        phone vocabulary file.
  --speaker-dict SPEAKER_DICT
                        speaker id map file for multiple speaker model.
  --test-metadata TEST_METADATA
                        test metadata.
  --output-dir OUTPUT_DIR
                        output dir.
  --ngpu NGPU           if ngpu == 0, use cpu.
  --verbose VERBOSE     verbose.

./local/synthesize_e2e.sh calls ${BIN_DIR}/synthesize_e2e.py, which can synthesize waveform from text file.

CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh ${conf_path} ${train_output_path} ${ckpt_name}

usage: synthesize_e2e.py [-h] [--fastspeech2-config FASTSPEECH2_CONFIG]
                         [--fastspeech2-checkpoint FASTSPEECH2_CHECKPOINT]
                         [--fastspeech2-stat FASTSPEECH2_STAT]
                         [--pwg-config PWG_CONFIG]
                         [--pwg-checkpoint PWG_CHECKPOINT]
                         [--pwg-stat PWG_STAT] [--phones-dict PHONES_DICT]
                         [--text TEXT] [--output-dir OUTPUT_DIR]
                         [--inference-dir INFERENCE_DIR] [--ngpu NGPU]
                         [--verbose VERBOSE]

Synthesize with fastspeech2 & parallel wavegan.

optional arguments:
  -h, --help            show this help message and exit
  --fastspeech2-config FASTSPEECH2_CONFIG
                        fastspeech2 config file.
  --fastspeech2-checkpoint FASTSPEECH2_CHECKPOINT
                        fastspeech2 checkpoint to load.
  --fastspeech2-stat FASTSPEECH2_STAT
                        mean and standard deviation used to normalize
                        spectrogram when training fastspeech2.
  --pwg-config PWG_CONFIG
                        parallel wavegan config file.
  --pwg-checkpoint PWG_CHECKPOINT
                        parallel wavegan generator parameters to load.
  --pwg-stat PWG_STAT   mean and standard deviation used to normalize
                        spectrogram when training parallel wavegan.
  --phones-dict PHONES_DICT
                        phone vocabulary file.
  --text TEXT           text to synthesize, a 'utt_id sentence' pair per line.
  --output-dir OUTPUT_DIR
                        output dir.
  --inference-dir INFERENCE_DIR
                        dir to save inference models
  --ngpu NGPU           if ngpu == 0, use cpu.
  --verbose VERBOSE     verbose.

1. --fastspeech2-config, --fastspeech2-checkpoint, --fastspeech2-stat and --phones-dict are arguments for fastspeech2, which correspond to the 4 files in the fastspeech2 pretrained model.
2. --pwg-config, --pwg-checkpoint, --pwg-stat are arguments for parallel wavegan, which correspond to the 3 files in the parallel wavegan pretrained model.
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.

Inferencing

After Synthesize, we will get static models of fastspeech2 and pwgan in ${train_output_path}/inference. ./local/inference.sh calls ${BIN_DIR}/inference.py, which provides a paddle static model inference example for fastspeech2 + pwgan synthesize.

CUDA_VISIBLE_DEVICES=${gpus} ./local/inference.sh ${train_output_path}

Pretrained Model

Pretrained FastSpeech2 model with no silence in the edge of audios:

• fastspeech2_nosil_baker_ckpt_0.4.zip
• fastspeech2_conformer_baker_ckpt_0.5.zip

Static model can be downloaded here fastspeech2_nosil_baker_static_0.4.zip.

Model	Step	eval/loss	eval/l1_loss	eval/duration_loss	eval/pitch_loss	eval/energy_loss
default	2(gpu) x 76000	1.0991	0.59132	0.035815	0.31915	0.15287
conformer	2(gpu) x 76000	1.0675	0.56103	0.035869	0.31553	0.15509

FastSpeech2 checkpoint contains files listed below.

fastspeech2_nosil_baker_ckpt_0.4
├── default.yaml            # default config used to train fastspeech2
├── phone_id_map.txt        # phone vocabulary file when training fastspeech2
├── snapshot_iter_76000.pdz # model parameters and optimizer states
└── speech_stats.npy        # statistics used to normalize spectrogram when training fastspeech2

You can use the following scripts to synthesize for ${BIN_DIR}/../sentences.txt using pretrained fastspeech2 and parallel wavegan models.

source path.sh

FLAGS_allocator_strategy=naive_best_fit \
FLAGS_fraction_of_gpu_memory_to_use=0.01 \
python3 ${BIN_DIR}/synthesize_e2e.py \
  --fastspeech2-config=fastspeech2_nosil_baker_ckpt_0.4/default.yaml \
  --fastspeech2-checkpoint=fastspeech2_nosil_baker_ckpt_0.4/snapshot_iter_76000.pdz \
  --fastspeech2-stat=fastspeech2_nosil_baker_ckpt_0.4/speech_stats.npy \
  --pwg-config=pwg_baker_ckpt_0.4/pwg_default.yaml \
  --pwg-checkpoint=pwg_baker_ckpt_0.4/pwg_snapshot_iter_400000.pdz \
  --pwg-stat=pwg_baker_ckpt_0.4/pwg_stats.npy \
  --text=${BIN_DIR}/../sentences.txt \
  --output-dir=exp/default/test_e2e \
  --inference-dir=exp/default/inference \
  --phones-dict=fastspeech2_nosil_baker_ckpt_0.4/phone_id_map.txt