Update app.py

app.py

@@ -207,8 +207,24 @@ def process_media(file_path, is_url=False):
         # Transcribe chunks
         transcription = transcribe_audio_chunks(chunks)
         
-        # Diarization (simplified as OpenAI doesn't provide speaker diarization)
-        formatted_transcript = f"Speaker 1: {transcription}"
+        # Diarization using OpenAI
+        diarization_prompt = f"""
+        The following is a transcription of a conversation. Please identify different speakers and label them as Speaker 1, Speaker 2, etc. Format the output as a series of speaker labels followed by their dialogue. Here's the transcription:
+
+        {transcription}
+
+        Please analyze the content and speaking styles to differentiate between speakers. Consider changes in topic, speaking patterns, and any contextual clues that might indicate a change in speaker.
+        """
+
+        diarization_response = openai.ChatCompletion.create(
+            model="gpt-3.5-turbo",
+            messages=[
+                {"role": "system", "content": "You are an AI assistant skilled in analyzing conversations and identifying different speakers."},
+                {"role": "user", "content": diarization_prompt}
+            ]
+        )
+
+        formatted_transcript = diarization_response['choices'][0]['message']['content']
         
         transcription_text = formatted_transcript
         generated_file = io.BytesIO(transcription_text.encode())