Skip to content

Commit 1f377a4

Browse files
Echo-NieEcho-Nie
authored
【PaddleSpeech No.13、14】补全合成系列中的脚本中参数缺失 (#4013)
* update examples/vctk/tts3/README.md * update examples/vctk/ernie_sat; examples/vctk/tts3 * update examples/vctk/ernie_sat; examples/vctk/tts3 * fix the errors found after the review
1 parent 290ce59 commit 1f377a4

File tree

4 files changed

+18
-10
lines changed

4 files changed

+18
-10
lines changed

‎examples/vctk/ernie_sat/README.md

Lines changed: 4 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -85,9 +85,12 @@ hifigan_vctk_ckpt_0.2.0
8585
```
8686
`./local/synthesize.sh` calls `${BIN_DIR}/../synthesize.py`, which can synthesize waveform from `metadata.jsonl`.
8787
```bash
88-
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh ${conf_path} ${train_output_path} ${ckpt_name}
88+
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name}
8989
```
90+
`--stage` controls the vocoder model during synthesis, which can be `0` , use`hifigan` model as vocoder.
91+
9092
## Speech Synthesis and Speech Editing
93+
9194
### Prepare
9295
**prepare aligner**
9396
```bash

‎examples/vctk/ernie_sat/run.sh

Lines changed: 4 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -27,10 +27,11 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
2727
fi
2828

2929
if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
30-
# synthesize, vocoder is pwgan
31-
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh ${conf_path} ${train_output_path} ${ckpt_name} || exit -1
30+
# synthesize, vocoder is hifigan by default
31+
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name} || exit -1
3232
fi
3333

3434
if [ ${stage} -le 3 ] && [ ${stop_stage} -ge 3 ]; then
35-
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh ${conf_path} ${train_output_path} ${ckpt_name} || exit -1
35+
# synthesize, task_name is speech synthesize by default stage 0, stage 1 will use speech edit as taskname
36+
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name} || exit -1
3637
fi

‎examples/vctk/tts3/README.md

Lines changed: 6 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -108,8 +108,10 @@ pwg_vctk_ckpt_0.1.1
108108
```
109109
`./local/synthesize.sh` calls `${BIN_DIR}/../synthesize.py`, which can synthesize waveform from `metadata.jsonl`.
110110
```bash
111-
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh ${conf_path} ${train_output_path} ${ckpt_name}
111+
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name}
112112
```
113+
`--stage` controls the vocoder model during synthesis, which can be `0` or `1`, use `pwgan` or `hifigan` model as vocoder.
114+
113115
```text
114116
usage: synthesize.py [-h]
115117
[--am {speedyspeech_csmsc,fastspeech2_csmsc,fastspeech2_ljspeech,fastspeech2_aishell3,fastspeech2_vctk,tacotron2_csmsc,tacotron2_ljspeech,tacotron2_aishell3}]
@@ -156,8 +158,10 @@ optional arguments:
156158
```
157159
`./local/synthesize_e2e.sh` calls `${BIN_DIR}/../synthesize_e2e.py`, which can synthesize waveform from text file.
158160
```bash
159-
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh ${conf_path} ${train_output_path} ${ckpt_name}
161+
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name}
160162
```
163+
`--stage` controls the vocoder model during synthesis, which can be `0` or `1`, use `pwgan` or `hifigan` model as vocoder.
164+
161165
```text
162166
usage: synthesize_e2e.py [-h]
163167
[--am {speedyspeech_csmsc,speedyspeech_aishell3,fastspeech2_csmsc,fastspeech2_ljspeech,fastspeech2_aishell3,fastspeech2_vctk,tacotron2_csmsc,tacotron2_ljspeech}]

‎examples/vctk/tts3/run.sh

Lines changed: 4 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -27,13 +27,13 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
2727
fi
2828

2929
if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
30-
# synthesize, vocoder is pwgan by default
31-
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh ${conf_path} ${train_output_path} ${ckpt_name} || exit -1
30+
# synthesize, vocoder is pwgan by default stage 0, stage 1 will use hifigan as vocoder
31+
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name} || exit -1
3232
fi
3333

3434
if [ ${stage} -le 3 ] && [ ${stop_stage} -ge 3 ]; then
35-
# synthesize_e2e, vocoder is pwgan by default
36-
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh ${conf_path} ${train_output_path} ${ckpt_name} || exit -1
35+
# synthesize_e2e, vocoder is pwgan by default 0, stage 1 will use hifigan as vocoder
36+
CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name} || exit -1
3737
fi
3838

3939
if [ ${stage} -le 4 ] && [ ${stop_stage} -ge 4 ]; then

0 commit comments

Comments
 (0)