Somewhat covered by other comments, but what I would want to see to evaluate a project like this is your current response latency and your goal latency for each step, or ideas about reducing it.
Show HN: Project S.A.T.U.R.D.A.Y. – open-source, self hosted, J.A.R.V.I.S.
31–32 of 32 posts
Re: Show HN: Project S.A.T.U.R.D.A.Y. – open-source, self hosted, J.A.R.V.I.S.
#32Somewhat covered by other comments, but what I would want to see to evaluate a project like this is your current response latency and your goal latency for each step, or ideas about reducing it.
Had a thought, maybe someone here will know the answer -- has anyone trained a voice system on how long to wait before replying to an utterance? It occured to me that if your STT hears "hey how's it going" it should reply immediately, but if it hears "I've been thinking about ." then it'd be appropriate to wait a second or two in case there's more coming.