From 7db18434bfd339b169f8a0b86882d58c73ed39cf Mon Sep 17 00:00:00 2001
From: Echo-Nie <157974576+Echo-Nie@users.noreply.github.com>
Date: Tue, 1 Apr 2025 13:34:04 +0800
Subject: [PATCH] =?UTF-8?q?=E3=80=90PaddleSpeech=20No.11=E3=80=91=E8=A1=A5?=
 =?UTF-8?q?=E5=85=A8=E5=90=88=E6=88=90=E7=B3=BB=E5=88=97=E4=B8=AD=E7=9A=84?=
 =?UTF-8?q?=E8=84=9A=E6=9C=AC=E4=B8=AD=E5=8F=82=E6=95=B0=E7=BC=BA=E5=A4=B1?=
 =?UTF-8?q?=20(#4038)?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 examples/ljspeech/tts3/README.md | 8 ++++++--
 examples/ljspeech/tts3/run.sh    | 8 ++++----
 2 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/examples/ljspeech/tts3/README.md b/examples/ljspeech/tts3/README.md
index cac214e74..12bd73777 100644
--- a/examples/ljspeech/tts3/README.md
+++ b/examples/ljspeech/tts3/README.md
@@ -105,8 +105,10 @@ pwg_ljspeech_ckpt_0.5
 ```
 `./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} ${ckpt_name}
+CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name}
 ```
+`--stage` controls the vocoder model during synthesis, which can be `0` or `1`, use `pwgan` or `hifigan` model as vocoder.
+
 ```text
 usage: synthesize.py [-h]
                      [--am {speedyspeech_csmsc,fastspeech2_csmsc,fastspeech2_ljspeech,fastspeech2_aishell3,fastspeech2_vctk,tacotron2_csmsc,tacotron2_ljspeech,tacotron2_aishell3}]
@@ -153,8 +155,10 @@ optional arguments:
 ```
 `./local/synthesize_e2e.sh` calls `${BIN_DIR}/../synthesize_e2e.py`, which can synthesize waveform from text file.
 ```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/ljspeech/tts3/run.sh b/examples/ljspeech/tts3/run.sh
index 0d8da920c..b02126e6e 100755
--- a/examples/ljspeech/tts3/run.sh
+++ b/examples/ljspeech/tts3/run.sh
@@ -27,13 +27,13 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
 fi
 
 if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then
-    # synthesize, vocoder is pwgan by default
-    CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh ${conf_path} ${train_output_path} ${ckpt_name} || exit -1
+    # synthesize, vocoder is pwgan by default stage 0, stage 1 will use hifigan as vocoder
+    CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name} || exit -1
 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} ${ckpt_name} || exit -1
+    # synthesize_e2e, vocoder is pwgan by default stage 0, stage 1 will use hifigan as vocoder
+    CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name} || exit -1
 fi
 
 if [ ${stage} -le 4 ] && [ ${stop_stage} -ge 4 ]; then