Update app.py

app.py

@@ -11,9 +11,8 @@ def pad_buffer(audio):
         audio = audio + b'\0' * (element_size - (buffer_size % element_size))
     return audio 
 
-def generate_voice(text):
+def generate_voice(text, voice_name):
     model_name = "eleven_multilingual_v1"
-    voice_name = "Antonio"
     try:
         audio = generate(
             text[:250],  # Limit to 250 characters
@@ -26,13 +25,27 @@ def generate_voice(text):
     except Exception as e:
         raise gr.Error(str(e))
 
+# Fetch all available voices
+all_voices = voices()
+
+# Filter out the desired voices and set Antonio as default
+desired_voices = ["Anita", "Rachael", "Michael", "Antonio"]
+filtered_voices = [voice.name for voice in all_voices if voice.name in desired_voices]
+
 input_text = gr.Textbox(
     label="Input Text (250 characters max)", 
     lines=2, 
-    value="Ciao, proviamo a dare una voce alle tue parole. Scrivi qui ciò che vorresti ascoltare!",
+    value="Hahaha OHH MY GOD! This is SOOO funny, I-I am Eleven a text-to-speech system!",
     elem_id="input_text"
 )
 
+input_voice = gr.Dropdown(
+    choices=filtered_voices, 
+    default="Antonio",
+    label="Voice", 
+    elem_id="input_voice"
+)
+
 out_audio = gr.Audio(
     label="Generated Voice",
     type="numpy", 
@@ -41,7 +54,7 @@ out_audio = gr.Audio(
 
 iface = gr.Interface(
     fn=generate_voice,
-    inputs=input_text,
+    inputs=[input_text, input_voice],
     outputs=out_audio,
     live=True,
     theme="Monochrome",