Following on from Save speech/voice into a local audio file…
My understanding is that Jibo won’t deliver raw audio and video to third-party devs due to privacy concerns, and will instead make feature data available from the raw video and audio.
If this is the case, here’s my plug for those of us hoping to do music related skills that the feature data extracted from the raw audio feed include some features useful to music skills. Offhand, I can think of beat detection, spectrum data, and pitch/timing info as being useful. With a bit of research or other folks chiming in, I’m sure we could flesh this out a bit.
From what I’ve seen in various responses from @Admin.JiboSupport, likes on a feature request help the SDK prioritize feature development, so if this would be useful to you, please slap a like on it. And if you have other music-related features that would be handy to detect, add them to the discussion.