add nr result note and correct python sample code error

Author: jimxiei

articles/cognitive-services/Speech-Service/includes/how-to/conversation-transcription/real-time-csharp.md

@@ -108,6 +108,9 @@ This sample code does the following:
 > [!NOTE]
 > `AudioStreamReader` is a helper class you can get on [GitHub](https://github.com/Azure-Samples/cognitive-services-speech-sdk/blob/master/quickstart/csharp/dotnet/conversation-transcription/helloworld/AudioStreamReader.cs).
 
+> [!NOTE]
+> If speaker identification or differentiate is enabled, even you have received `Transcribed` results, service is still evaluating them by accumulated audio information. When service finds any previous result assigned a wrong `UserId`, it will send out an almost same `Transcribed` result again, but only the `UserId` and `UtteranceId` are different. Since the `UtteranceId` format is `{index}_{UserId}_{Offset}`, when you receive a `Transcribed` result, you could use `UtteranceId` to determine if current `Transcribed` result is going to correct previous one. Your client or UI logic could decide behaviors, like overwriting previous output, or just ignore the current one.
+
 Call the function `TranscribeConversationsAsync()` to start conversation transcription.
 
 ```csharp

articles/cognitive-services/Speech-Service/includes/how-to/conversation-transcription/real-time-javascript.md

@@ -74,6 +74,9 @@ This sample code does the following:
 * Registers to events and begins transcription.
 * If you want to differentiate speakers without providing voice samples, please enable `DifferentiateGuestSpeakers` feature as in [Conversation Transcription Overview](../../../conversation-transcription.md). 
 
+> [!NOTE]
+> If speaker identification or differentiate is enabled, even you have received `transcribed` results, service is still evaluating them by accumulated audio information. When service finds any previous result assigned a wrong `speakerId`, it will send out an almost same `transcribed` result again, but only the `speakerId` and `UtteranceId` are different. Since the `UtteranceId` format is `{index}_{speakerId}_{Offset}`, when you receive a `transcribed` result, you could use `UtteranceId` to determine if current `transcribed` result is going to correct previous one. Your client or UI logic could decide behaviors, like overwriting previous output, or just ignore the current one.
+
 ```javascript
 (function() {
     "use strict";

articles/cognitive-services/Speech-Service/includes/how-to/conversation-transcription/real-time-python.md

@@ -75,6 +75,9 @@ This sample code does the following:
 * Read the whole wave files at once and stream it to sdk and begins transcription.
 * If you want to differentiate speakers without providing voice samples, please enable `DifferentiateGuestSpeakers` feature as in [Conversation Transcription Overview](../../../conversation-transcription.md). 
 
+> [!NOTE]
+> If speaker identification or differentiate is enabled, even you have received `transcribed` results, service is still evaluating them by accumulated audio information. When service finds any previous result assigned a wrong `speakerId`, it will send out an almost same `transcribed` result again, but only the `speakerId` and `UtteranceId` are different. Since the `UtteranceId` format is `{index}_{speakerId}_{Offset}`, when you receive a `transcribed` result, you could use `UtteranceId` to determine if current `transcribed` result is going to correct previous one. Your client or UI logic could decide behaviors, like overwriting previous output, or just ignore the current one.
+
 ```python
 import azure.cognitiveservices.speech as speechsdk
 import time
@@ -123,8 +126,8 @@ def conversation_transcription_differentiate_speakers():
     user1 = speechsdk.transcription.Participant("[email protected]", "en-us", voice_signature_user1)
     user2 = speechsdk.transcription.Participant("[email protected]", "en-us", voice_signature_user2)
 
-    conversation.add_participant_async(user1)
-    conversation.add_participant_async(user2)
+    conversation.add_participant_async(user1).get()
+    conversation.add_participant_async(user2).get()
     transcriber.join_conversation_async(conversation).get()
     transcriber.start_transcribing_async()