【PaddleSpeech No.12】 (#4037)

Echo-Nie · web-flow · commit fae44e7e78e2 · 2025-04-07T15:15:40.000+08:00
* fix the svs1

* del stage 0

* &#21024;&#21435;&#20013;&#25991;README&#20013;&#22810;&#20313;&#30340;stage 0
diff --git a/examples/opencpop/svs1/README.md b/examples/opencpop/svs1/README.md
@@ -118,6 +118,8 @@ pwgan_opencpop_ckpt_1.4.0.zip
 ```bash
 CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh ${conf_path} ${train_output_path} ${ckpt_name}
 ```
+use `pwgan` model as vocoder.
+
 ```text
 usage: synthesize.py [-h]
                      [--am {diffsinger_opencpop}]
@@ -170,8 +172,10 @@ optional arguments:
 `local/pinyin_to_phone.txt` comes from the readme of the opencpop dataset, indicating the mapping from pinyin to phonemes in opencpop.
 
 ```bash
-CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh ${conf_path} ${train_output_path} ${ckpt_name}
+CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name}
 ```
+`--stage` controls the vocoder model during synthesis, which can be `0` or `1`, use `pwgan` or `hifigan` model as vocoder.
+
 ```text
 usage: synthesize_e2e.py [-h]
                          [--am {speedyspeech_csmsc,speedyspeech_aishell3,fastspeech2_csmsc,fastspeech2_ljspeech,fastspeech2_aishell3,fastspeech2_vctk,tacotron2_csmsc,tacotron2_ljspeech}]
diff --git a/examples/opencpop/svs1/README_cn.md b/examples/opencpop/svs1/README_cn.md
@@ -121,6 +121,8 @@ pwgan_opencpop_ckpt_1.4.0.zip
 ```bash
 CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh ${conf_path} ${train_output_path} ${ckpt_name}
 ```
+&#20351;&#29992; `pwgan` &#27169;&#22411;&#20316;&#20026;&#22768;&#30721;&#22120;&#12290;
+
 ```text
 usage: synthesize.py [-h]
                      [--am {diffsinger_opencpop}]
@@ -173,8 +175,10 @@ optional arguments:
 `local/pinyin_to_phone.txt`&#26469;&#28304;&#20110;opencpop&#25968;&#25454;&#38598;&#20013;&#30340;README&#65292;&#34920;&#31034;opencpop&#20013;&#25340;&#38899;&#21040;&#38899;&#32032;&#30340;&#26144;&#23556;&#12290;
 
 ```bash
-CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh ${conf_path} ${train_output_path} ${ckpt_name}
+CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name}
 ```
+`--stage` &#29992;&#20110;&#36873;&#25321;&#21512;&#25104;&#26102;&#20351;&#29992;&#30340;&#22768;&#30721;&#22120;&#27169;&#22411;&#65292;&#21462;&#20540;&#20026; `0` &#25110; `1`&#65292;&#20998;&#21035;&#23545;&#24212;&#20351;&#29992; `pwgan` &#25110; `hifigan` &#27169;&#22411;&#20316;&#20026;&#22768;&#30721;&#22120;&#12290;
+
 ```text
 usage: synthesize_e2e.py [-h]
                          [--am {speedyspeech_csmsc,speedyspeech_aishell3,fastspeech2_csmsc,fastspeech2_ljspeech,fastspeech2_aishell3,fastspeech2_vctk,tacotron2_csmsc,tacotron2_ljspeech}]
diff --git a/examples/opencpop/svs1/run.sh b/examples/opencpop/svs1/run.sh
@@ -32,6 +32,6 @@ if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
 fi
 
 if [ ${stage} -le 3 ] && [ ${stop_stage} -ge 3 ]; then
-    # synthesize_e2e, vocoder is pwgan by default
-    CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh ${conf_path} ${train_output_path} ${ckpt_name} || exit -1
+    # synthesize_e2e, vocoder is pwgan by default, stage 1 will use hifigan as vocoder
+    CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name} || exit -1
 fi
