Actions
Forum Thread Details
  • State Suggested Answer
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 31 subscribers
  • Views 734 views
  • Users 0 members are here
Related

Connecting 3.5mm jack to logic board in place of surface mount mic.

adminindisguise

I am trying to find out if this is even possible. Been watching Ben Heck talk about how great this community is for over a year now and thought it would be a great place to find an answer.

 

I am working on a raspberry pi project where I will make an Echo Dot controllable via voice chat in a Discord server.

 

The jist of it is that I want to do is connect the audio out from the raspberry pi to the microphone connection on the Echo Dot in the place of a surface mount mic.

From there I will connect the audio out from the dot back into the pi, but that will be simple seeing as how there is already an audio out on the dot.

 

My question is: Would it even be possible to run that 3.5mm audio signal into a surface mount input.

It makes sense in my head that if I make the output mono I should be able to use either the L or R connection coming from the 3.5mm into the input mount for the mic. But before I go and destroy a $50 Dot I just wanted to know if it is theoretically possible or if I'm wasting my time.

 

Thanks much for any information you can give!

  • 0

    Hi Blake,

    The one problem that I can see, and there may be others , is that the signal levels and impedance will not match and your audio may be garbled by over driving the input where the mic was.

    John

  • 0

    Hi Blake,

    I agree with John, and there was also some teardown of the Echo dot online, and it contains multiple microphones. It may misbehave if only one or two are connected up, because it is possibly running some algorithms to aid in speech processing.

    Knowing Amazon there is possibly some API that could be used instead of wiring into the mics, i.e. bypass the Echo dot altogether. However it would then be a software exercise to interwork the voice chat into the Amazon API.

    EDIT: looks like there is an API, available in UK/Germany in early 2017, I don't know if it is available now or later in other regions.

  • 0 in reply to shabaz

    I have seen some of the API related projects and know that it is a possible solution; but one major aspect of my project is the desire to control it entirely via speech. The software to detect the trigger word is written directly into the device and is not available as part of the API that I have seen so far, so I would need to find some second voice recognition module and that just sounds like a mess. Plus, in my experience with echosim.com the API is far slower than the actual device when making inquiries.

     

    The microphones are only so the device can pinpoint the direction that the command is coming from so it can target that source and drown out the other sounds in a room. Very neat technology, but I don't believe that would mess with the devices ability to hear the command if there was only one input. Probably going to give it a shot and see what happens, it's a relatively inexpensive project and even if it doesn't work I could probably get the dot back together and working afterwards.

     

    Thanks very much for your input!

  • 0 in reply to jw0752

    I'm wondering if there may be a way to use the Pi's GPIO pins for a lower power audio out, or possibly adding some resistance between the units. I am fairly inexperienced with altering signals and don't know how that would play out. Thanks for your insight on the project. I'll keep updating this thread if I go through with it and let everyone know how it goes.

  • 0 in reply to adminindisguise

    Hi Blake,

     

    If you can find out the output level and know the input level yes the signal can be attenuated with resistors and capacitors.

     

    John