User avatar
bonzadog
Posts: 215
Joined: Wed Apr 25, 2012 9:40 am
Location: Rietberg/Germany

Snore to non-snore detector.

Sun Dec 24, 2017 2:50 pm

24.12.2017
What seems like a rather strange title has a very serious background.

I an looking for a contactless method of detecting Sleep apnea - this is when breathing stops for a while
during sleep.
So one need to detect a snoore ends then time the it takes for the start of next snore - This time would be then checked against a
value for an alarm.
The main problem is what mictophone to use , how to define a snoore start and stop....ie no false readings though cars/motorbikes etc.

I realise this a has it drawbacks but The person involved dislikes being connected to any devices...so a pure acoustic method must be used.

Any advice on this please?

mrs
Posts: 3
Joined: Wed Dec 13, 2017 4:35 pm

Re: Snore to non-snore detector.

Sat Dec 30, 2017 8:49 pm

Hi!

This is an ambitious project. I definitely think it is possible, but you will need time.

Hardware-wise an easy solution would be the Audioinjector soundcard, which sports a build-in microphone [1].
Software-wise, I would encourage you to check out the free speech recognition toolkit Kaldi [2].
It can be trained to recognize audible sounds.
You will first need to record data and annotate it.
Then, you can use the annotated samples to train and test the recognition system.

Best wishes,
Marc

[1] http://www.audioinjector.net/rpi-hat
[2] http://kaldi-asr.org/

User avatar
bonzadog
Posts: 215
Joined: Wed Apr 25, 2012 9:40 am
Location: Rietberg/Germany

Re: Snore to non-snore detector.

Sun Dec 31, 2017 3:05 pm

Hello Marc,

Many thanks for those tips. I will most certainly try that.
This project is very important for me.

BD

shadetree01010100
Posts: 29
Joined: Thu Oct 12, 2017 9:19 pm
Location: CO, USA
Contact: Website

Re: Snore to non-snore detector.

Mon Jan 29, 2018 4:23 pm

This sensor breakout isn't too expensive, and may simplify the task of quantifying sound in general: https://www.sparkfun.com/products/12642

Return to “Assistive technology and accessibility”