Update app.py

app.py

@@ -8,46 +8,25 @@ import torch
 from transformers import pipeline
 import librosa
 import gradio as gr
-import requests  # Added for debugging network
+import requests
 
-# HugChat login credentials from environment variables (secrets)
-EMAIL = os.environ.get("Email")
-PASSWD = os.environ.get("Password")
-
-# Debug: Print credentials to verify they're being read
-print("EMAIL from env:", EMAIL)
-print("PASSWORD from env:", PASSWD)
-
-# Directory to store cookies
+# Directory to store/load cookies
 cookie_path_dir = "./cookies/"
 os.makedirs(cookie_path_dir, exist_ok=True)
 
-# Test network connectivity to Hugging Face
-try:
-    response = requests.get("https://huggingface.co/login", timeout=10)
-    print("Network test: Successfully reached https://huggingface.co/login, status code:", response.status_code)
-except Exception as e:
-    print("Network test failed:", str(e))
-
-# Login to HugChat with detailed error handling
+# Load pre-saved cookies instead of logging in
 try:
-    sign = Login(EMAIL, PASSWD)
-    print("Attempting login with hugchat...")
-    cookies = sign.login(cookie_dir_path=cookie_path_dir, save_cookies=True)
-    print("Login successful, cookies obtained.")
-    chatbot = hugchat.ChatBot(cookies=cookies.get_dict())
+    print("Attempting to load cookies from:", cookie_path_dir)
+    chatbot = hugchat.ChatBot(cookie_path_dir=cookie_path_dir)
+    print("Cookies loaded successfully.")
 except Exception as e:
-    print(f"Login failed with error: {str(e)}")
-    print("Full traceback:")
-    import traceback
-    traceback.print_exc()
+    print(f"Failed to load cookies: {str(e)}")
     sys.exit(1)
 
 # Model and device configuration for Whisper transcription
 MODEL_NAME = "openai/whisper-large-v3-turbo"
 device = 0 if torch.cuda.is_available() else "cpu"
 
-# Initialize Whisper pipeline
 pipe = pipeline(
     task="automatic-speech-recognition",
     model=MODEL_NAME,
@@ -56,9 +35,6 @@ pipe = pipeline(
 )
 
 def transcribe_audio(audio_path):
-    """
-    Transcribe a local audio file using the Whisper pipeline.
-    """
     try:
         audio, sr = librosa.load(audio_path, sr=16000, mono=True)
         transcription = pipe(audio, batch_size=8, generate_kwargs={"language": "urdu"})["text"]
@@ -67,9 +43,6 @@ def transcribe_audio(audio_path):
         return f"Error processing audio: {e}"
 
 def extract_metadata(file_name):
-    """
-    Extract metadata from the file name.
-    """
     base = file_name.split(".")[0]
     parts = base.split("_")
     if len(parts) >= 3:
@@ -80,9 +53,6 @@ def extract_metadata(file_name):
     return {"agent_username": "Unknown", "location": "Unknown"}
 
 def process_audio(audio, file_name):
-    """
-    Process the audio file and return Urdu transcription, English translation, and crops with diseases.
-    """
     urdu_text = transcribe_audio(audio)
     if "Error" in urdu_text:
         return json.dumps({"error": urdu_text})
@@ -180,7 +150,6 @@ def process_audio(audio, file_name):
 
     return json.dumps(output)
 
-# Gradio Interface
 with gr.Blocks(title="Audio to Crop Disease API") as interface:
     gr.Markdown("## Upload Audio to Get Urdu Transcription, English Translation, and Crop Diseases")