I’m wondering if anyone has had any experience with getting live audio and having it translated in real-time (or close to real-time) as subtitles/text directly into TouchDesigner? I’ve thought about live transcription/speech-to-text models that then feed into translation models, however this seems like it could get clunky/messy very quickly.
For context, there will be a presenter speaking in English, and then viewers will be sitting in front of a touchscreen where they would be able to turn on translation for their language.
Any tips or pointers in any direction (whether the software/techniques are paid or free) would be greatly appreciated!