Update app.py

app.py

@@ -5,6 +5,9 @@ from PIL import Image
 from io import BytesIO
 import os
 from elevenlabs.client import ElevenLabs
+import requests
+import cv2
+import numpy as np
 
 # Set API keys as environment variables
 os.environ["TOGETHER_API_KEY"] = st.secrets['together_api']
@@ -69,7 +72,7 @@ def tts(text):
                 f.write(chunk)
         
         # Play the audio in Streamlit
-        st.audio(audio_file_path, format="audio/mp3",autoplay=True)
+        st.audio(audio_file_path, format="audio/mp3", autoplay=True)
     except Exception as e:
         st.error(f"Error generating speech: {e}")
 
@@ -126,24 +129,38 @@ st.sidebar.markdown("## About")
 st.sidebar.markdown("This app uses advanced AI to describe what it sees through your camera in real-time.")
 st.sidebar.markdown("Powered by [Together AI](https://together.ai) and Streamlit.")
 
-# Access the camera
-img_file_buffer = st.camera_input("Take a picture")
-
-if img_file_buffer is not None:
-    
-    # Convert the image file buffer to a PIL Image
-    img = Image.open(img_file_buffer)
-
-    # Display the captured image
-    st.image(img, caption='Captured Image', width=300)
-
-    # Get and display the description
-    with st.spinner('🔍 Analyzing the image...'):
-        description = get_image_description(img)
-    st.success('✅ Analysis complete!')
-    st.markdown("### AI Description:")
-    st.write(description)
-
-    # Convert description to speech and play it
-    
-    tts(description)
+# IP Webcam URL input
+st.markdown("### Connect to IP Webcam")
+ip_webcam_url = 'https://192.168.31.68:8080'
+
+if ip_webcam_url:
+    # Fetch the video stream from IP Webcam
+    video_url = f"{ip_webcam_url}/video"
+    st.markdown(f"### Live Video Stream from IP Webcam")
+    st.image(video_url, use_column_width=True)
+
+    # Capture a frame from the video stream
+    if st.button("Capture Image"):
+        try:
+            # Fetch the latest frame from the video stream
+            response = requests.get(f"{ip_webcam_url}/shot.jpg", stream=True)
+            if response.status_code == 200:
+                # Convert the response content to a PIL Image
+                img = Image.open(BytesIO(response.content))
+
+                # Display the captured image
+                st.image(img, caption='Captured Image', width=300)
+
+                # Get and display the description
+                with st.spinner('🔍 Analyzing the image...'):
+                    description = get_image_description(img)
+                st.success('✅ Analysis complete!')
+                st.markdown("### AI Description:")
+                st.write(description)
+
+                # Convert description to speech and play it
+                tts(description)
+            else:
+                st.error("Failed to fetch image from IP Webcam.")
+        except Exception as e:
+            st.error(f"Error capturing image: {e}")