Update app2.py

app2.py

@@ -63,6 +63,7 @@ def create_vc_fn(model_name, tgt_sr, net_g, vc, if_f0, version, file_index):
         vc_upload,
         tts_text,
         tts_voice,
+	      tts_rate,
         f0_up_key,
         f0_method,
         index_rate,
@@ -95,7 +96,13 @@ def create_vc_fn(model_name, tgt_sr, net_g, vc, if_f0, version, file_index):
                     return "Text is too long", None
                 if tts_text is None or tts_voice is None:
                     return "You need to enter text and select a voice", None
-                asyncio.run(edge_tts.Communicate(tts_text, "-".join(tts_voice.split('-')[:-1])).save("tts.mp3"))
+                inc_rate = "+0%"
+                if tts_rate < 0 :
+                  inc_rate = (f"{100 - tts_rate}%")
+                else:
+                  inc_rate = (f"+{tts_rate - s100}%")
+
+                asyncio.run(edge_tts.Communicate(text=tts_text, voice= "-".join(tts_voice.split('-')[:-1]), rate= "-80%").save("tts.mp3"))
                 audio, sr = librosa.load("tts.mp3", sr=16000, mono=True)
                 vc_input = "tts.mp3"
             times = [0, 0, 0]
@@ -337,7 +344,8 @@ def change_audio_mode(vc_audio_mode):
             gr.Button.update(visible=False),
             # TTS
             gr.Textbox.update(visible=False),
-            gr.Dropdown.update(visible=False)
+            gr.Dropdown.update(visible=False),
+            gr.Number.update(visible=False)
         )
     elif vc_audio_mode == "Upload audio":
         return (
@@ -363,7 +371,8 @@ def change_audio_mode(vc_audio_mode):
             gr.Button.update(visible=False),
             # TTS
             gr.Textbox.update(visible=False),
-            gr.Dropdown.update(visible=False)
+            gr.Dropdown.update(visible=False),
+            gr.Number.update(visible=False)
         )
     elif vc_audio_mode == "Youtube":
         return (
@@ -389,7 +398,8 @@ def change_audio_mode(vc_audio_mode):
             gr.Button.update(visible=True),
             # TTS
             gr.Textbox.update(visible=False),
-            gr.Dropdown.update(visible=False)
+            gr.Dropdown.update(visible=False),
+            gr.Number.update(visible=False)
         )
     elif vc_audio_mode == "TTS Audio":
         return (
@@ -415,7 +425,8 @@ def change_audio_mode(vc_audio_mode):
             gr.Button.update(visible=False),
             # TTS
             gr.Textbox.update(visible=True),
-            gr.Dropdown.update(visible=True)
+            gr.Dropdown.update(visible=True),
+            gr.Textbox.update(visible=True)
         )
 
 def use_microphone(microphone):
@@ -630,6 +641,7 @@ if __name__ == '__main__':
                                                 # TTS
                                                 tts_text = gr.Textbox(label="TTS text", info="Text to speech input", visible=False)
                                                 tts_voice = gr.Dropdown(label="Edge-tts speaker", choices=voices, visible=False, allow_custom_value=False, value="en-US-AnaNeural-Female")
+                                                tts_rate = gr.Textbox(label="TTS Rate", info='Change tts output speed (100 = normal)')
                                             with gr.Column():
                                                 vc_split_model = gr.Dropdown(label="Splitter Model", choices=["hdemucs_mmi", "htdemucs", "htdemucs_ft", "mdx", "mdx_q", "mdx_extra_q"], allow_custom_value=False, visible=False, value="htdemucs", info="Select the splitter model (Default: htdemucs)")
                                                 vc_split_log = gr.Textbox(label="Output Information", visible=False, interactive=False)
@@ -877,7 +889,8 @@ if __name__ == '__main__':
                                 vc_combined_output,
                                 vc_combine,
                                 tts_text,
-                                tts_voice
+                                tts_voice,
+				tts_rate
                             ]
                         )
         # Audio tool