From ca03f4db214b7300a00fd7159ee3a5b0a8fada91 Mon Sep 17 00:00:00 2001
From: Echo-Nie <157974576+Echo-Nie@users.noreply.github.com>
Date: Tue, 18 Mar 2025 14:50:37 +0800
Subject: [PATCH] =?UTF-8?q?=E3=80=90PaddleSpeech=20No.6=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(#4004)?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

* run.sh修改：为 synthesize 和 synthesize_e2e 添加 --stage 参数控制 vocoder 模型选择，REAMDE.md修改：补充 stage 参数说明，明确 vocoder 选择逻辑

* 添加run.sh中stage参数相关的注释

* HiFiGAN改为MultiBand MelGAN

* cmsc文件改回原位（No.15不修改），这里只对No.6做修改
---
 examples/canton/tts3/README.md | 1 +
 examples/canton/tts3/run.sh    | 8 ++++----
 2 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/examples/canton/tts3/README.md b/examples/canton/tts3/README.md
index 87ef40907..d2c46f642 100644
--- a/examples/canton/tts3/README.md
+++ b/examples/canton/tts3/README.md
@@ -37,6 +37,7 @@ Run the command below to
 3. train the model.
 4. synthesize wavs.
     - synthesize waveform from `metadata.jsonl`.
+    - `--stage` controls the vocoder model during synthesis (0 = pwgan, 1 = hifigan).
     - synthesize waveform from text file.
 ```bash
 ./run.sh
diff --git a/examples/canton/tts3/run.sh b/examples/canton/tts3/run.sh
index acfc50223..0e1f52a1c 100755
--- a/examples/canton/tts3/run.sh
+++ b/examples/canton/tts3/run.sh
@@ -28,13 +28,13 @@ 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