sync ms

app.py

@@ -54,7 +54,7 @@ def get_text(text, language_str, hps):
     return bert, phone, tone, language
 
 
-def infer(text, sdp_ratio, noise_scale, noise_scale_w, length_scale, sid):
+def TTS_infer(text, sdp_ratio, noise_scale, noise_scale_w, length_scale, sid):
     global net_g
     bert, phones, tones, lang_ids = get_text(text, "ZH", hps)
     with torch.no_grad():
@@ -88,7 +88,7 @@ def infer(text, sdp_ratio, noise_scale, noise_scale_w, length_scale, sid):
 
 def tts_fn(text, speaker, sdp_ratio, noise_scale, noise_scale_w, length_scale):
     with torch.no_grad():
-        audio = infer(
+        audio = TTS_infer(
             text,
             sdp_ratio=sdp_ratio,
             noise_scale=noise_scale,
@@ -121,23 +121,14 @@ def concatenate_audios(audio_samples, sample_rate=44100):
 
 
 def read_text(file_path: str):
-    try:
-        # 打开文件并读取内容
-        with open(file_path, "r", encoding="utf-8") as file:
-            content = file.read()
-            return content
-
-    except FileNotFoundError:
-        print(f"文件未找到: {file_path}")
-
-    except IOError:
-        print(f"读取文件时发生错误: {file_path}")
-
-    except Exception as e:
-        print(f"发生未知错误: {e}")
+    with open(file_path, "r", encoding="utf-8") as file:
+        content = file.read()
+        return content
 
 
-def infer_tab1(text, speaker, sdp_ratio, noise_scale, noise_scale_w, length_scale):
+def infer_upl(text, speaker, sdp_ratio, noise_scale, noise_scale_w, length_scale):
+    status = "Success"
+    audio = content = None
     try:
         content = read_text(text)
         sentences = text_splitter(content)
@@ -145,7 +136,7 @@ def infer_tab1(text, speaker, sdp_ratio, noise_scale, noise_scale_w, length_scal
         for sentence in tqdm(sentences, desc="TTS 推理中..."):
             with torch.no_grad():
                 audios.append(
-                    infer(
+                    TTS_infer(
                         sentence,
                         sdp_ratio=sdp_ratio,
                         noise_scale=noise_scale,
@@ -155,20 +146,24 @@ def infer_tab1(text, speaker, sdp_ratio, noise_scale, noise_scale_w, length_scal
                     )
                 )
 
-        return concatenate_audios(audios, hps.data.sampling_rate), content
+        audio = concatenate_audios(audios, hps.data.sampling_rate)
 
     except Exception as e:
-        return None, f"{e}"
+        status = f"{e}"
 
+    return status, audio, content
 
-def infer_tab2(content, speaker, sdp_ratio, noise_scale, noise_scale_w, length_scale):
+
+def infer_txt(content, speaker, sdp_ratio, noise_scale, noise_scale_w, length_scale):
+    status = "Success"
+    audio = None
     try:
         sentences = text_splitter(content)
         audios = []
         for sentence in tqdm(sentences, desc="TTS 推理中..."):
             with torch.no_grad():
                 audios.append(
-                    infer(
+                    TTS_infer(
                         sentence,
                         sdp_ratio=sdp_ratio,
                         noise_scale=noise_scale,
@@ -178,11 +173,12 @@ def infer_tab2(content, speaker, sdp_ratio, noise_scale, noise_scale_w, length_s
                     )
                 )
 
-        return concatenate_audios(audios, hps.data.sampling_rate)
+        audio = concatenate_audios(audios, hps.data.sampling_rate)
 
     except Exception as e:
-        print(f"{e}")
-        return None
+        status = f"{e}"
+
+    return status, audio
 
 
 if __name__ == "__main__":
@@ -222,7 +218,7 @@ if __name__ == "__main__":
 
         with gr.Tab(_L("输入模式")):
             gr.Interface(
-                fn=infer_tab2,  # 使用 text_to_speech 函数
+                fn=infer_txt,  # 使用 text_to_speech 函数
                 inputs=[
                     gr.TextArea(
                         label=_L("请输入简体中文文案"),
@@ -251,14 +247,17 @@ if __name__ == "__main__":
                         minimum=0.1, maximum=2, value=1, step=0.1, label=_L("生成时长")
                     ),
                 ],
-                outputs=gr.Audio(label=_L("输出音频")),
+                outputs=[
+                    gr.Textbox(label=_L("状态栏"), show_copy_button=True),
+                    gr.Audio(label=_L("输出音频")),
+                ],
                 flagging_mode="never",
                 concurrency_limit=4,
             )
 
         with gr.Tab(_L("上传模式")):
             gr.Interface(
-                fn=infer_tab1,  # 使用 text_to_speech 函数
+                fn=infer_upl,  # 使用 text_to_speech 函数
                 inputs=[
                     gr.components.File(
                         label=_L("请上传简体中文 TXT 文案"),
@@ -288,6 +287,7 @@ if __name__ == "__main__":
                     ),
                 ],
                 outputs=[
+                    gr.Textbox(label=_L("状态栏"), show_copy_button=True),
                     gr.Audio(label=_L("输出音频")),
                     gr.TextArea(label=_L("文案提取结果"), show_copy_button=True),
                 ],