TianYuan
3 years ago
commit
0bc9450c51
|
Before Width: | Height: | Size: 229 KiB |
|
After Width: | Height: | Size: 1.5 MiB |
File diff suppressed because one or more lines are too long
@ -0,0 +1,146 @@
|
||||
# Parallel WaveGAN with AISHELL-3
|
||||
This example contains code used to train a [parallel wavegan](http://arxiv.org/abs/1910.11480) model with [AISHELL-3](http://www.aishelltech.com/aishell_3).
|
||||
|
||||
AISHELL-3 is a large-scale and high-fidelity multi-speaker Mandarin speech corpus which could be used to train multi-speaker Text-to-Speech (TTS) systems.
|
||||
## Dataset
|
||||
### Download and Extract the datasaet
|
||||
Download AISHELL-3.
|
||||
```bash
|
||||
wget https://www.openslr.org/resources/93/data_aishell3.tgz
|
||||
```
|
||||
Extract AISHELL-3.
|
||||
```bash
|
||||
mkdir data_aishell3
|
||||
tar zxvf data_aishell3.tgz -C data_aishell3
|
||||
```
|
||||
### Get MFA result of AISHELL-3 and Extract it
|
||||
We use [MFA2.x](https://github.com/MontrealCorpusTools/Montreal-Forced-Aligner) to get durations for aishell3_fastspeech2.
|
||||
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 own MFA model reference to [use_mfa example](https://github.com/PaddlePaddle/DeepSpeech/tree/develop/examples/other/use_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`.
|
||||
```bash
|
||||
./run.sh
|
||||
```
|
||||
### Preprocess the dataset
|
||||
```bash
|
||||
./local/preprocess.sh ${conf_path}
|
||||
```
|
||||
When it is done. A `dump` folder is created in the current directory. The structure of the dump folder is listed below.
|
||||
|
||||
```text
|
||||
dump
|
||||
├── dev
|
||||
│ ├── norm
|
||||
│ └── raw
|
||||
├── test
|
||||
│ ├── norm
|
||||
│ └── raw
|
||||
└── train
|
||||
├── norm
|
||||
├── raw
|
||||
└── feats_stats.npy
|
||||
```
|
||||
|
||||
The dataset is split into 3 parts, namely `train`, `dev` and `test`, each of which contains a `norm` and `raw` subfolder. The `raw` folder contains log magnitude of mel spectrogram of each utterances, while the norm folder contains normalized spectrogram. The statistics used to normalize the spectrogram is computed from the training set, which is located in `dump/train/feats_stats.npy`.
|
||||
|
||||
Also there is a `metadata.jsonl` in each subfolder. It is a table-like file which contains id and paths to spectrogam of each utterance.
|
||||
|
||||
### Train the model
|
||||
```bash
|
||||
CUDA_VISIBLE_DEVICES=${gpus} ./local/train.sh ${conf_path} ${train_output_path}
|
||||
```
|
||||
`./local/train.sh` calls `${BIN_DIR}/train.py`.
|
||||
Here's the complete help message.
|
||||
|
||||
```text
|
||||
usage: train.py [-h] [--config CONFIG] [--train-metadata TRAIN_METADATA]
|
||||
[--dev-metadata DEV_METADATA] [--output-dir OUTPUT_DIR]
|
||||
[--ngpu NGPU] [--verbose VERBOSE] [--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.
|
||||
--verbose VERBOSE verbose.
|
||||
|
||||
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 save in `checkpoints/` inside this directory.
|
||||
4. `--ngpu` is the number of gpus to use, if ngpu == 0, use cpu.
|
||||
|
||||
### Synthesize
|
||||
`./local/synthesize.sh` calls `${BIN_DIR}/synthesize.py`, which can synthesize waveform from `metadata.jsonl`.
|
||||
```bash
|
||||
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh ${conf_path} ${train_output_path} ${ckpt_name}
|
||||
```
|
||||
```text
|
||||
usage: synthesize.py [-h] [--config CONFIG] [--checkpoint CHECKPOINT]
|
||||
[--test-metadata TEST_METADATA] [--output-dir OUTPUT_DIR]
|
||||
[--ngpu NGPU] [--verbose VERBOSE]
|
||||
|
||||
Synthesize with parallel wavegan.
|
||||
|
||||
optional arguments:
|
||||
-h, --help show this help message and exit
|
||||
--config CONFIG parallel wavegan 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.
|
||||
--verbose VERBOSE verbose.
|
||||
```
|
||||
|
||||
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. If you use the pretrained model, use the `snapshot_iter_1000000.pdz `.
|
||||
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 Models
|
||||
Pretrained models can be downloaded here [pwg_aishell3_ckpt_0.5.zip](https://paddlespeech.bj.bcebos.com/Parakeet/pwg_aishell3_ckpt_0.5.zip).
|
||||
|
||||
Parallel WaveGAN checkpoint contains files listed below.
|
||||
|
||||
```text
|
||||
pwg_aishell3_ckpt_0.5
|
||||
├── default.yaml # default config used to train parallel wavegan
|
||||
├── feats_stats.npy # statistics used to normalize spectrogram when training parallel wavegan
|
||||
└── snapshot_iter_1000000.pdz # generator parameters of parallel wavegan
|
||||
```
|
||||
## Acknowledgement
|
||||
We adapted some code from https://github.com/kan-bayashi/ParallelWaveGAN.
|
||||
@ -1,12 +0,0 @@
|
||||
python analysis.py \
|
||||
--filename "recoder_mp_bs16_fp32_ngpu8.txt" \
|
||||
--keyword "ips[sent./sec]:" \
|
||||
--base_batch_size 16 \
|
||||
--model_name "Conformer" \
|
||||
--mission_name "eight gpu" \
|
||||
--run_mode "mp" \
|
||||
--ips_unit "sent./sec" \
|
||||
--gpu_num 8 \
|
||||
--use_num 480 \
|
||||
--separator " " \
|
||||
--direction_id "1"
|
||||
@ -1,12 +0,0 @@
|
||||
python analysis.py \
|
||||
--filename "recoder_sp_bs16_fp32_ngpu1.txt" \
|
||||
--keyword "ips[sent./sec]:" \
|
||||
--base_batch_size 16 \
|
||||
--model_name "Conformer" \
|
||||
--mission_name "one gpu" \
|
||||
--run_mode "sp" \
|
||||
--ips_unit "sent./sec" \
|
||||
--gpu_num 1 \
|
||||
--use_num 60 \
|
||||
--separator " " \
|
||||
--direction_id "1"
|
||||
@ -0,0 +1,4 @@
|
||||
#!/bin/bash
|
||||
|
||||
test -d montreal-forced-aligner || wget --no-check-certificate https://github.com/MontrealCorpusTools/Montreal-Forced-Aligner/releases/download/v1.0.1/montreal-forced-aligner_linux.tar.gz
|
||||
tar xvf montreal-forced-aligner_linux.tar.gz
|
||||
@ -0,0 +1,26 @@
|
||||
#!/bin/bash
|
||||
|
||||
WGET="wget --no-check-certificate"
|
||||
|
||||
# SCTK official repo does not have version tags. Here's the mapping:
|
||||
# # 2.4.9 = 659bc36; 2.4.10 = d914e1b; 2.4.11 = 20159b5.
|
||||
SCTK_GITHASH=20159b5
|
||||
SCTK_CXFLAGS="-w -march=native"
|
||||
CFLAGS="CFLAGS=${SCTK_CXFLAGS}"
|
||||
CXXFLAGS="CXXFLAGS=-std=c++11 ${SCTK_CXFLAGS}"
|
||||
|
||||
MAKE=make
|
||||
|
||||
|
||||
${WGET} -nv -T 10 -t 3 -O sctk-${SCTK_GITHASH}.tar.gz https://github.com/usnistgov/SCTK/archive/${SCTK_GITHASH}.tar.gz;
|
||||
tar zxvf sctk-${SCTK_GITHASH}.tar.gz
|
||||
rm -rf sctk-${SCTK_GITHASH} sctk
|
||||
mv SCTK-${SCTK_GITHASH}* sctk-${SCTK_GITHASH}
|
||||
ln -s sctk-${SCTK_GITHASH} sctk
|
||||
touch sctk-${SCTK_GITHASH}.tar.gz
|
||||
|
||||
rm -f sctk/.compiled
|
||||
CFLAGS="${SCTK_CXFLAGS}" CXXFLAGS="-std=c++11 ${SCTK_CXFLAGS}" ${MAKE} -C sctk config
|
||||
CFLAGS="${SCTK_CXFLAGS}" CXXFLAGS="-std=c++11 ${SCTK_CXFLAGS}" ${MAKE} -C sctk all doc
|
||||
${MAKE} -C sctk install
|
||||
touch sctk/.compiled
|
||||
@ -0,0 +1,6 @@
|
||||
#!/bin/bash
|
||||
|
||||
apt install -y libvorbis-dev libmp3lame-dev libmad-ocaml-dev
|
||||
test -d sox-14.4.2 || wget --no-check-certificate https://nchc.dl.sourceforge.net/project/sox/sox/14.4.2/sox-14.4.2.tar.gz
|
||||
tar -xvzf sox-14.4.2.tar.gz -C .
|
||||
cd sox-14.4.2 && ./configure --prefix=/usr/ && make -j4 && make install
|
||||
@ -0,0 +1,36 @@
|
||||
|
||||
#!/bin/bash
|
||||
# copy from Espnet
|
||||
|
||||
set -euo pipefail
|
||||
|
||||
if [ $# -eq 0 ] || [ $# -gt 2 ]; then
|
||||
echo "Usage: $0 <python> [venv-path]"
|
||||
echo "e.g."
|
||||
echo "$0 \$(which python3)"
|
||||
exit 1;
|
||||
elif [ $# -eq 2 ]; then
|
||||
PYTHON="$1"
|
||||
VENV="$2"
|
||||
elif [ $# -eq 1 ]; then
|
||||
PYTHON="$1"
|
||||
VENV="venv"
|
||||
fi
|
||||
|
||||
if ! "${PYTHON}" -m venv --help > /dev/null 2>&1; then
|
||||
echo "Error: ${PYTHON} is not Python3?"
|
||||
exit 1
|
||||
fi
|
||||
if [ -e activate_python.sh ]; then
|
||||
echo "Warning: activate_python.sh already exists. It will be overwritten"
|
||||
fi
|
||||
|
||||
"${PYTHON}" -m venv ${VENV}
|
||||
cat << EOF > activate_python.sh
|
||||
#!/usr/bin/env bash
|
||||
# THIS FILE IS GENERATED BY tools/setup_venv.sh
|
||||
. $(cd ${VENV}; pwd)/bin/activate
|
||||
EOF
|
||||
|
||||
. ./activate_python.sh
|
||||
python3 -m pip install -U pip wheel
|
||||
Loading…
Reference in new issue