Create predtiction.py to get the Predictions

predictions.py

@@ -0,0 +1,24 @@
+from PIL import Image, ImageDraw
+from helper import summarize_predictions_natural_language,render_results_in_image
+from transformers import pipeline
+
+# Load object detection pipeline
+object_detection_pipe = pipeline("object-detection", model="facebook/detr-resnet-50")
+
+# Load text-to-speech pipeline
+tts_pipe = pipeline("text-to-speech", model="kakao-enterprise/vits-ljs")
+
+def get_predictions(uploaded_image):
+    pil_image = Image.open(uploaded_image)
+
+    # Perform object detection
+    pipeline_output = object_detection_pipe(pil_image)
+    processed_image = render_results_in_image(pil_image, pipeline_output)
+
+    # Summarize predictions
+    text = summarize_predictions_natural_language(pipeline_output)
+
+    # Generate audio from text
+    narrated_text = tts_pipe(text)
+
+    return processed_image, text, narrated_text["audio"][0]