I would like to send generated audio from a text-to-speech plugin via Sound Submix to an external Python application (pybinsim), which performs specialized 3D rendering. It’s important that only the generated audio is streamed to the application, as ambience and FX audio should be rendered on the Unreal side. I have already tried using FMOD and WWISE via AudioLink, but they do not allow external forwarding of the audio stream. Does anyone have a suggestion on how to implement such audio streaming?
The simplest option to write audio from unreal to some file on disk would be using the metasound wav writer node, but if you’d like to capture a full submix this might not be what you’re after, essentially with a little c++ you could write a custom submix effect class that writes the audio buffers it receives to a file or passes them to some external library, which sounds like what you might be after, without c++ though I think that the wav writer might be the only option.