I am exploring the possibility of streaming real-time audio from an active Architect flow to an external biometrics API. I am looking at using a Data Action to post the audio data. Does the platform support a streaming buffer via Data Actions or is there another endpoint for raw audio?
Data Actions are not for streaming audio. They are for JSON payloads. If you want to do biometrics, you should look at the Audio Connector. It is built exactly for what you are trying to do. It streams the audio over a websocket.
This is exactly the kind of thing that makes me want to pull my hair out. Data Actions are for small data lookups, not media streams. That is right, you have to use Audio Connector.
Setting it up is a massive pain because you have to manage the websocket server yourself, but it is the only way the platform allows real-time audio access.