WIP

app.py

@@ -214,7 +214,9 @@ def check_api_health():
         
         if cpu_percent > 90 or memory_percent > 90:
             logger.warning(f"High resource usage detected - CPU: {cpu_percent}%, Memory: {memory_percent}%")
-            
+        
+        logger.info("API health check passed successfully")
+
         return True
         
     except requests.exceptions.RequestException as e:
@@ -262,10 +264,17 @@ def transcribe_youtube(url, return_timestamps, generate_subs, chunk_length_s=30,
         metadata = result.get("metadata", {})
         logger.info(f"Transcription metadata: {metadata}")
         
-        # Format response with segments
+        # Format response with segments (without id)
         formatted_result = {
             "text": result["transcription"]["text"],
-            "segments": result["transcription"]["segments"] if return_timestamps else None
+            "segments": [
+                {
+                    "start": segment["start"],
+                    "end": segment["end"],
+                    "text": segment["text"]
+                }
+                for segment in result["transcription"]["segments"]
+            ] if return_timestamps else None
         }
         
         # Generate subtitles if requested
@@ -303,6 +312,8 @@ def transcribe(inputs, return_timestamps, generate_subs, chunk_length_s=30, batc
                 "chunk_length_s": chunk_length_s,
                 "batch_size": batch_size
             }
+
+            logger.info(f"Sending request to API with parameters: {params}")
             
             # Send request to API
             response = requests.post(
@@ -317,10 +328,17 @@ def transcribe(inputs, return_timestamps, generate_subs, chunk_length_s=30, batc
             metadata = result.get("metadata", {})
             logger.info(f"Transcription metadata: {metadata}")
             
-            # Format response with segments
+            # Format response with segments (without id)
             formatted_result = {
                 "text": result["transcription"]["text"],
-                "segments": result["transcription"]["segments"] if return_timestamps else None
+                "segments": [
+                    {
+                        "start": segment["start"],
+                        "end": segment["end"],
+                        "text": segment["text"]
+                    }
+                    for segment in result["transcription"]["segments"]
+                ] if return_timestamps else None
             }
             
             # Generate subtitles if requested
@@ -417,3 +435,4 @@ with demo:
 logger.info("Starting Gradio interface")
 demo.queue().launch(ssr_mode=False)
 
+