### Answer
Vapi does support recording calls, which inherently includes both the bot's and the user's audio. However, the documentation does not explicitly detail the functionality to split these audio recordings by user and bot directly through the API [(1)](
) .
For your specific need to separate audio tracks, you might consider implementing additional processing after you retrieve the full call recording. This could involve using audio processing tools to differentiate and split the tracks based on voice recognition or other audio signatures.