Update app.py

app.py

@@ -1,6 +1,8 @@
 import gradio as gr
 import torch
 import torchaudio
+import os
+import subprocess
 from transformers import AutoModelForCausalLM
 from outetts.wav_tokenizer.decoder import WavTokenizer
 from yarngpt.audiotokenizer import AudioTokenizer
@@ -22,6 +24,22 @@ def initialize_model():
     
     return model, audio_tokenizer
 
+    def download_if_not_exists(url, filename):
+    if not os.path.exists(filename):
+        print(f"Downloading {filename}...")
+        subprocess.run(["wget", url, "-O", filename])
+        print(f"Downloaded {filename}")
+
+# Download the model files if they're not already present
+download_if_not_exists(
+    "https://huggingface.co/novateur/WavTokenizer-medium-speech-75token/resolve/main/wavtokenizer_mediumdata_frame75_3s_nq1_code4096_dim512_kmeans200_attn.yaml",
+    "wavtokenizer_config.yaml"
+)
+download_if_not_exists(
+    "https://huggingface.co/novateur/WavTokenizer-large-speech-75token/blob/main/wavtokenizer_large_speech_320_v2.ckpt",
+    "wavtokenizer_model.ckpt"
+)
+
 # Generate audio from text
 def generate_speech(text, speaker_name):
     # Create prompt