Update app.py

app.py

@@ -8,32 +8,18 @@ zh_tts = TTS(model_name="tts_models/zh-CN/baker/tacotron2-DDC-GST", progress_bar
 de_tts = TTS(model_name="tts_models/de/thorsten/vits", gpu=False)
 es_tts = TTS(model_name="tts_models/es/mai/tacotron2-DDC", progress_bar=False, gpu=False)
 
-def text_to_speech(text: str, speaker_wav, speaker_wav_file, language: str):
-    if speaker_wav_file and not speaker_wav:
-        speaker_wav = speaker_wav_file
-    file_path = "output.wav"
-    if language == "zh-CN":
-        # if speaker_wav is not None:
-        #     zh_tts.tts_to_file(text, speaker_wav=speaker_wav, file_path=file_path)
-        # else:
-        zh_tts.tts_to_file(text, file_path=file_path)
-    elif language == "de":
-        # if speaker_wav is not None:
-        #     de_tts.tts_to_file(text, speaker_wav=speaker_wav, file_path=file_path)
-        # else:
-        de_tts.tts_to_file(text, file_path=file_path)
-    elif language == "es":
-        # if speaker_wav is not None:
-        #     es_tts.tts_to_file(text, speaker_wav=speaker_wav, file_path=file_path)
-        # else:
-        es_tts.tts_to_file(text, file_path=file_path)
-    else:
+def text_to_speech(text: str, speaker_wav, speaker_wav_file):
+    if len(text) > 0:
+        if speaker_wav_file and not speaker_wav:
+            speaker_wav = speaker_wav_file
+        file_path = "output.wav"
         if speaker_wav is not None:
-            tts.tts_to_file(text, speaker_wav=speaker_wav, language=language, file_path=file_path)
+            tts.tts_to_file(text, speaker_wav=speaker_wav, language="en", file_path=file_path)
         else:
-            tts.tts_to_file(text, speaker=tts.speakers[0], language=language, file_path=file_path)
-    return file_path
-
+            tts.tts_to_file(text, speaker=tts.speakers[0], language="en", file_path=file_path)
+        return file_path
+    else:
+        raise gr.Error("Input box cannot be blank")
 
 
 title = "Voice-Cloning-Demo"
@@ -44,27 +30,15 @@ def toggle(choice):
     else:
         return gr.update(visible=False, value=None), gr.update(visible=True, value=None)
 
-def handle_language_change(choice):
-    if choice == "zh-CN" or choice == "de" or choice == "es":
-        return gr.update(visible=False), gr.update(visible=False), gr.update(visible=False)
-    else:
-        return gr.update(visible=True), gr.update(visible=True), gr.update(visible=True)
-
-def btn_input(text_input):
-    if len(text_input) > 0:
-        new_btn = gr.Button("Submit", variant="primary", interactive=True)
-    else:
-        new_btn = gr.Button("Submit", variant="primary", interactive=False)
-    return new_btn
+# def btn_input(text_input):
+#     if len(text_input) == 0:
+#         raise gr.Error("Input box cannot be blank")
 
-warming_text = """Please note that Chinese, German, and Spanish are currently not supported for voice cloning."""
 
 with gr.Blocks() as demo:
     with gr.Row():
         with gr.Column():
             text_input = gr.Textbox(label="Input the text", value="", max_lines=3)
-            lan_input = gr.Radio(label="Language", choices=["en", "fr-fr", "pt-br", "zh-CN", "de", "es"], value="en")
-            gr.Markdown(warming_text)
             radio = gr.Radio(["mic", "file"], value="mic",
                              label="How would you like to upload your audio?")
             audio_input_mic = gr.Audio(label="Voice to clone", sources="microphone", type="filepath", visible=True)
@@ -74,16 +48,14 @@ with gr.Blocks() as demo:
                 with gr.Column():
                     btn_clear = gr.Button("Clear")
                 with gr.Column():
-                    btn = gr.Button("Submit", variant="primary", interactive=False)
+                    btn = gr.Button("Submit", variant="primary")
         with gr.Column():
             audio_output = gr.Audio(label="Output")
 
     # gr.Examples(examples, fn=inference, inputs=[audio_file, text_input],
     #                   outputs=audio_output, cache_examples=True)
-    btn.click(text_to_speech, inputs=[text_input, audio_input_mic,
-              audio_input_file, lan_input], outputs=audio_output)
+    btn.click(text_to_speech, inputs=[text_input, audio_input_mic,audio_input_file], outputs=audio_output)
     radio.change(toggle, radio, [audio_input_mic, audio_input_file])
-    lan_input.change(handle_language_change, lan_input, [radio, audio_input_mic, audio_input_file])
-    text_input.change(btn_input, text_input, btn)
+    # btn.click(btn_input, inputs=text_input, outputs=None)
 
 demo.launch(share=True)