Why is Reasoning/Thinking now included in Text-to-Speech playback since the latest updates?

[ronniekolk]

Since the latest update(s), I noticed that when using Text-to-Speech (TTS) in LibreChat, the reasoning or "thinking" part of the AI's response is also read aloud, not just the main answer.

Was this an intentional change as part of the new Reasoning UI and Chain-of-Thought features, or is it a side effect of recent refactoring?

Previously, only the main answer was included in TTS, but now the reasoning seems to be a distinct part of the message content and is picked up by TTS as well.

Could you clarify if this was an intended design change, and if so, what is the best way to exclude the reasoning/thinking part from being read aloud by TTS if a user prefers that?