Is anyone on this forum acquainted with the technology employed to analyse audio signal for signs of depression.
I have 3 years worth of daily voice memos which I would be interested in exploring to see if my development of depression and anxiety could be discernible on differences in my vocal tone.
If anyone has any pointers I would be most grateul
I tried some apps that claimed to detect mood from voice last year, but the results seemed a bit random… There is a lot of research in this space, and several companies are even selling products (including some for call centers!), but in order to track mental health. you’d probably need to 1. record your voice all day long, rather than relying on a short sample, and 2. factor in other data, like how much you talk, and how active you are etc…
If there was an app that simply extracts a few basic metrics from your voice (frequency range? pauses?), rather than trying to reach any conclusions, I’d be interested, too!
Thank you - I’ve looked into the OSS software used in some of the literature used in vocal biomarker studies. I will dig them out, I can’t recall off the top of my head whether it required a large durational sample of ‘regular’ voice data to compare against.
I’d like to work on how possible that is. I believe that any open stack of self-tracking needs to not cede voice analysis to proprietary platforms when a lot can be inferred from it.
several companies are even selling products (including some for call centers!), but in order to track mental health.
OMG yes, I think I’ve seen one of those! I’m pretty sure https://www.cogitocorp.com/ started out as an app to help detect vocal traces of depression.