Last night I slept with a (cheap as chips) digital audio recorder on my nightstand, and have just taken a quick look/listen at/to the seven hour recording in Fission, a Mac audio editor that displays a waveform. Super-interesting! (What on earth was I shouting out about, 47 minutes in???)
I’d love to know if others have already walked this way, and in particular if anyone’s had thoughts about semi-automating analysis? For example — but certainly not limited to — tracking the snoring wavelength.
I use the iOS app SnoreLab to track snoring. Happy with the results. I believe you can have it record the full night instead of just samples. Their intent is for you to be able to play it back to your doctor for follow up.
I’ve considered recording my sleep with an audio device, but haven’t ever gotten around to it. It certainly would be rather interesting to hear my mumbling at 3:00 AM. Maybe you could find the key voice events from your audio recording and plot those on a timeline for each night; eventually finding a pattern or three within the data. Could be interesting?
I’ve seen this a lot in subjective vs measured sleep, but never with audio. Sleep experts who have spoken at QS conferences tend to argue that people sleep more than they think, which makes sense: we remember the wake ups but forget the sleep. But I’ve never seen this explored with audio.
I use oura, Apple Watch 6 with AutoSleep and Garmin 945 to track sleep. I used an extra Arlo 3 HD camera for two nights trained on myself and found I move 10 to 15+ times a night even on nights with one very short awakening. It also seems to be a good way to double check when are deeply asleep.
How these compare 3 trackers is another thread but Oura and Apple seemed to be the best with audiovisual and self report as gold standard.