----------------------------- **While I generate an answer, here are some sources that might be helpful**: * [Introduction]() * [Implementing Realtime Conversation]() * [Orchestration Models]() * [Core Models]() -----------------------------

### Answer Currently, the Vapi documentation does not provide specific guidance or methods for dynamically changing the audio output device during an ongoing conversation directly through the Vapi platform [(1)]() . This functionality generally pertains to the client-side management of audio devices, which can depend greatly on the operating system and the environment (web, iOS, Android) in which your application is running. For web applications, managing audio output devices can typically be handled through the Web Audio API and the Media Devices API provided by modern browsers. You would need to implement this control within your application logic, not directly through Vapi. If you are developing for iOS or Android, similar native APIs are available that allow for the selection of audio output sources. These would also need to be controlled via the native development environment and not through Vapi directly. For recording the video with the end-of-call downloadable feature, ensure that your implementation captures both the audio and video streams during the call. Vapi itself focuses on managing the voice interactions, and does not directly handle video recording or changing system-level settings like audio output devices. If you're looking for more detailed control over the call and media handling, reviewing the capabilities of the client environments (like HTML5 APIs for web, or specific iOS/Android APIs) will be necessary.

Got it. Tagging @User for additional help.