Addresses #119 - update whisperx syntax to match latest version

Author: rishikanthc

diarize.py

@@ -37,7 +37,7 @@ def diarize_transcript(audio_file, transcript, device="cpu", model_name="pyannot
                 print("Could not import torch for CUDA memory management")
 
         print(f"Loading diarization model: {model_name}")
-        diarize_model = whisperx.DiarizationPipeline(
+        diarize_model = whisperx.diarize.DiarizationPipeline(
             model_name=model_name,
             device=device,
             use_auth_token=hf_token if hf_token else None
@@ -127,4 +127,4 @@ def diarize_transcript(audio_file, transcript, device="cpu", model_name="pyannot
         # Fallback: Assign "unknown" to original segments
         for segment in transcript["segments"]:
             segment["speaker"] = "unknown"
-        return transcript
+        return transcript