finally done maybe part 2

app.py

@@ -1,7 +1,7 @@
 import gradio as gr
 import torch
 import torchaudio
-from transformers import pipeline, AutoModel
+from transformers import pipeline, AutoModel, AutoConfig
 import librosa
 import numpy as np
 import re
@@ -17,6 +17,15 @@ from functools import lru_cache
 from enum import Enum
 from huggingface_hub import login, InferenceClient
 
+# Pre-load onnxruntime to handle stack execution issues
+try:
+    import onnxruntime as ort
+    logger_temp = logging.getLogger("onnxruntime_check")
+    logger_temp.info(f"✅ onnxruntime loaded successfully: {ort.__version__}")
+except Exception as e:
+    logger_temp = logging.getLogger("onnxruntime_check")
+    logger_temp.warning(f"⚠️ onnxruntime import issue: {e}")
+
 # ============================================
 # ENVIRONMENT & LOGGING SETUP
 # ============================================

packages.txt

@@ -0,0 +1,2 @@
+ffmpeg
+libsndfile1

requirements.txt

@@ -27,7 +27,7 @@ protobuf==3.20.3
 
 # ML Acceleration
 accelerate==0.25.0
-onnxruntime==1.16.3
+onnxruntime==1.17.1
 
 # Video/Media Processing
 torchcodec==0.1.0