feat: update to eleven multilingual v2

app.py

@@ -11,12 +11,12 @@ def pad_buffer(audio):
         audio = audio + b'\0' * (element_size - (buffer_size % element_size))
     return audio 
 
-def generate_voice(text, voice_name, model_name):
+def generate_voice(text, voice_name):
     try:
         audio = generate(
             text[:250], # Limit to 250 characters
             voice=voice_name, 
-            model=model_name
+            model="eleven_multilingual_v2"
         )
         return (44100, np.frombuffer(pad_buffer(audio), dtype=np.int16))
     except UnauthenticatedRateLimitError as e:
@@ -46,36 +46,29 @@ badges = """
 """
 
 description = """
-A demo of the world's most advanced TTS systems, made by [ElevenLabs](https://elevenlabs.io). Eleven Monolingual is designed to generate highly realistic voices in English, where Eleven Multilingual is a single model supporting multiple languages including English, German, Polish, Spanish, Italian, French, Portuguese, and Hindi. Sign up on [ElevenLabs](https://elevenlabs.io) to get fast access, long-form generation, voice cloning, API keys, and more!
+A demo of the world's most advanced TTS systems, made by [ElevenLabs](https://elevenlabs.io). Eleven Multilingual v2 is a single foundational model supporting 28 languages languages including English, Chinese, Spanish, Hindi, Portuguese, French, German, Japanese, Arabic, Korean, Indonesian, Italian, Dutch, Turkish, Polish, Swedish, Filipino, Malay, Romanian, Ukrainian, Greek, Czech, Danish, Finnish, Bulgarian, Croatian, Slovak, and Tamil. Sign up on [ElevenLabs](https://elevenlabs.io) to get fast access, long-form generation, voice cloning, API keys, and more!
 """
 
 with gr.Blocks() as block:
-    gr.Markdown('[ ![ElevenLabs](https://user-images.githubusercontent.com/12028621/243114574-21267d89-5e82-4e7e-9c81-caf30b237683.png) ](https://elevenlabs.io)')
+    gr.Markdown('[ ![ElevenLabs](https://user-images.githubusercontent.com/12028621/262629275-4f85c9cf-85b6-435e-ab50-5b8c7c4e9dd2.png) ](https://elevenlabs.io)')
     gr.Markdown(badges)
     gr.Markdown(description)
     
     input_text = gr.Textbox(
         label="Input Text (250 characters max)", 
         lines=2, 
-        value="Hahaha OHH MY GOD! This is SOOO funny, I-I am Eleven a text-to-speech system!",
+        value="Hello! 你好! Hola! नमस्ते! Olá! Bonjour! Hallo! こんにちは! مرحبا! 안녕하세요! Halo! Ciao! Hallo! Merhaba! Cześć! Hej! Kamusta! Halo! Bună! Привіт! Γειά σας! Ahoj! Hej! Hei! Здравей! Pozdrav! Ahoj! வணக்கம்!",
         elem_id="input_text"
     )
 
     all_voices = voices() 
     input_voice = gr.Dropdown(
         [ voice.name for voice in all_voices ], 
-        value="Arnold",
+        value="Bella",
         label="Voice", 
         elem_id="input_voice"
     )
 
-    input_model = gr.Radio(
-        ["eleven_monolingual_v1", "eleven_multilingual_v1"],
-        label="Model",
-        value="eleven_monolingual_v1",
-        elem_id="input_model",
-    )
-
     run_button = gr.Button(
         text="Generate Voice", 
         type="button"
@@ -87,7 +80,7 @@ with gr.Blocks() as block:
         elem_id="out_audio"
     )
         
-    inputs = [input_text, input_voice, input_model]
+    inputs = [input_text, input_voice]
     outputs = [out_audio]
     
     run_button.click(