Go Down

Topic: Is there an affordable way to recognise a sound and ping a phone? (Read 254 times) previous topic - next topic

Potatobob

Hi, I have a really basic knowledge of Arduino and programming thus I might need a bit more help to get started and I apologize for this.

This being said what I wish to do is create an affordable tool that recognizes a sound and that can ping a phone via an application or texto what ever is the simplest.

I know for sure that this can be achieved with Arduino but I would like to know what are my other options as I would like to sell the product afterwards and make profits.


I anticipate that my best option will be to ask a professional in electromechanic to create the tool for me.


So here are my questions I suppose:
1- Is it worth to design my tool with Arduino first and then contact an electromechanic professional to create the final product for me or it is a waste of time to design on Arduino to begin with since the final product and the Arduino version will not use the code nor the same components?
2- Do you already know of a product that can recognize a sound and that can ping a phone when it occurs?
3- Using Arduino, what are the components I need to create my tool? (I think a microphone, the Arduino board, maybe a breadboard, some wires and something for the Bluetooth or whatever can be used to ping a phone)

Thank you very much, any support is appreciated!

robfocus

Do you mean recognise a specific sound, or just any sound?

I think recognising a specific sound might be beyond the processing capabilities of Arduinos, unless the classification is in very simple terms like volume and frequency, and there are very cheap sensors for sound such this: https://www.adafruit.com/product/3421

In order to send a notification to a phone you'll probably need a GSM module (or GSM shield) and a sim-card, then the code to send a text message to a phone is very simple, there are good libraries for doing this. Otherwise you might be able to use a bluetooth module but the range would be limited.

As for the first question, I guess it depends on your budget and whether you want to learn from the experience of designing and making it. The Arduinos are really just ATmega microprocessors with a few convenient additions to regulate power, make programming them easy, and so on, but it's possible to strip all of that away to make a streamlined product so Arduino is probably a great way of developing a project of this kind.

Personally I'm a long time computer programmer but new to electronics, and it's proving to be very rewarding experience trying to make similar things to what you have in mind, so I would say go for it, there are lots of resources online and people willing to help.

(edit: for saying GPS when I meant GSM!!)

Potatobob

Thank you for your help that is exactly the information I was looking for.

My goal in fact is to allow users to record a sound so the device will be able to recognize that sound when it ears it.

I first tough I could make this in a very cheap manner but it makes sense that the processing power required is just too much. My second option is to use better components to make a higher quality product but that also has more features. By example it should be able to recognize that sound through other sounds playing.

Here is a straight forward example, I would like to record the microwave beeping sound and when ever the device ears the microwave beeping it messages my phone. But there would have been issues with cheap components anyway since it must be harder to recognize a sound in an active environment (lets say music is playing next to the device and the microwave).

So yeah thank you again and I think I will have a lot of researches and code to do to get there but I will begin by making sure this is a viable project because I also design video games and that is my core hobby!

Southpark

Here is a straight forward example, I would like to record the microwave beeping sound and when ever the device ears the microwave beeping it messages my phone.
This probably gets into the area of sound analysis and machine learning --- using computer analysis methods that process recorded sounds, and if the analysis results indicates that the probability of a section of analysed audio matching what you want --- then the computer can then be programmed to do something --- such as to tell the user it's detected something.

If you're focusing on electronic applicances turning on or off, then there are other ways to skin a cat (and eat them too ---- but of course, I'm only kidding here). For example -- there can be power monitoring devices that could be programmed to do something (like txt your phone etc.) whenever it senses a significant amount of current or power being used .... such as when the microwave oven turns on. This wouldn't be based on sound though. So if your idea is only focusing on sound --- then it will likely require sofware and hardware and machine learning and/or spectral analysis techniques to be applied.

Potatobob

Quote
This probably gets into the area of sound analysis and machine learning
I learnt a bit about neural networks and types of machine learning so I am confident that this is the way to go for my project. At this point I have a good Idea of what I want and how to achieve it but I think it will be a real pain to put that up together XD.

If I manage just to put all the right components together I will clearly feel accomplished but I am nowhere near expecting the device and software to work along and I think that balancing the machine learning in itself will be quite a challenge.


Thank you both for your support and I will surely come back to post my results if I do any kind of work on this device.

Quote
If you're focusing on electronic applicances turning on or off, then there are other ways to skin a cat
Well here is the thing:
My goal is not to create a device that will modify anything about the electronic devices at home but what it should do is simplify your life in various ways. By example you should be able to place the device in the kitchen and record a few sounds that are distinct such as the oven timer going off, the alarm system, the fire detector and so on.   This would allow for an extra layer of security at home, by example I want to go watch tv while the pizza is cooking well I can't because I will not ear the oven alarm, well this is solved as my oven will call me on my phone :P

My girlfriend asked about using this to know if a baby wakes up, and I totally agree, the current devices on the market to detect if your baby is crying are all speakers from his bed to your room, that is good but you have to carry the darn thing all over the place. Well not anymore !!!! Shebam you can simply have your phone in your pocket and as soon as the baby makes a sound you get pinged where ever you are in the home.

I am certain that the applications for this are infinite and that someone creative enough can make this real and make ton of money or just help making the world better.


Anyway, I have a lot of things going on in my life so I don't think I will personally create this device although I had the idea but if someone reading this feels like he can do it well go ahead m8! I am not going to stop you, I am not going to patent this any time soon and I would prefer seeing my ideas on the market even if someone else took the credit then never seeing my ideas working at all.

Fractoggen

Arduino is a prototype platform based on an easy tonuse hardware and software. It consists of a circuit board, which can be programmed and a ready made software called Arduino IDE, which is used to write and upload the computer code to the physical board.  :)

wvmarle

Recognising a specific sound is the hardest part. Sounds never sound exactly alike twice: there's always interference from other noises, differences in the exact way it's made, etc. That makes it really hard.

It depends also of course on WHAT sound you want to recognise - an Arduino can quite well recognise a sound that's much louder than the normal sound level: a clap sensor.

But if it's recognising a specific word as it's said in a regular conversation, that's a task even humans have difficulty with.
Quality of answers is related to the quality of questions. Good questions will get good answers. Useless answers are a sign of a poor question.

Go Up