moddified app.py

app.py

@@ -11,11 +11,11 @@ emotion_mapping = {"amused": 0, "anger": 1,
                    "disgust": 2, "neutral": 3, "sleepiness": 4}
 
 
-def synthesize_speech(input_type, text, speaker_id, embed_type, emotion_id):
+def synthesize_speech(input_type, text, own_text, speaker_id, embed_type, emotion_id):
     if input_type == "Choose from examples":
         selected_text = text
     else:
-        selected_text = input_type
+        selected_text = own_text
 
     if embed_type == "bert_embed":
         command = f"python3 synthesize.py --text '{selected_text}' --bert_embed 1 --speaker_id {speaker_id} --restore_step 900000 --mode single -p config/EmoV_DB/preprocess.yaml -m config/EmoV_DB/model.yaml -t config/EmoV_DB/train.yaml"
@@ -34,7 +34,7 @@ iface = gr.Interface(
             ["Choose from examples", "Enter your own text"], label="Input Type"),
         gr.inputs.Dropdown(choices=predefined_texts, label="Select a text"),
         gr.inputs.Textbox(lines=2, label="Enter your own text"),
-        gr.inputs.Slider(minimum=0, maximum=10, step=1,
+        gr.inputs.Slider(minimum=0, maximum=3, step=1,
                          default=0, label="Speaker ID"),
         gr.inputs.Radio(["bert_embed", "emotion_id"], label="Embedding Type"),
         gr.inputs.Dropdown(choices=emotion_mapping, label="Select Emotion"),
@@ -42,4 +42,4 @@ iface = gr.Interface(
     outputs=gr.outputs.Audio(type="filepath"),
     title="Text-to-Speech Demo",
 )
-iface.launch()
+iface.launch(share=True)