Initial

app.py

@@ -33,7 +33,8 @@ async def main(audio_file_path: str):
 
     Returns
     -------
-    None
+    dict
+        final_output
     """
     # Paths
     config_nemo = "config/nemo/diar_infer_telephonic.yaml"
@@ -73,7 +74,7 @@ async def main(audio_file_path: str):
     # Step 1: Detect Dialogue
     has_dialogue = dialogue_detector.process(audio_file_path)
     if not has_dialogue:
-        return
+        return {"error": "No dialogue detected in the audio file."}
 
     # Step 2: Speech Enhancement
     audio_path = enhancer.enhance_audio(
@@ -167,7 +168,7 @@ async def main(audio_file_path: str):
     )
     annotator.add_topic(topic_result)
 
-    #  Step 15: File/Audio Feature Extraction
+    # Step 15: File/Audio Feature Extraction
     props = audio_feature_extractor.properties()
 
     (
@@ -195,7 +196,7 @@ async def main(audio_file_path: str):
 
     final_output = annotator.finalize()
 
-    # Step 16: Tocal Silence Calculation
+    # Step 16: Total Silence Calculation
     stats = SilenceStats.from_segments(final_output['ssm'])
     t_std = stats.threshold_std(factor=0.99)
     final_output["silence"] = t_std
@@ -269,6 +270,7 @@ async def main(audio_file_path: str):
     # Step 18: Clean Up
     cleaner.cleanup(temp_dir, audio_file_path)
 
+    return final_output
 
 async def process(path: str):
     """
@@ -289,4 +291,5 @@ async def process(path: str):
 
 if __name__ == "__main__":
     directory_to_watch = ".data/input"
-    Watcher.start_watcher(directory_to_watch, process)
+    # Watcher.start_watcher(directory_to_watch, process)
+    pass