• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 100 subscribers
  • Views 273 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Direction location by using stereo audio

Bob Morris
Intellectual 355 points
Other Parts Discussed in Thread: TMS320C6748, TMS320C6424, TMS320C6657

I have a new project, and I'd appreciate some direction to get started. What would it take to locate a sound source based on two microphone inputs, like we do all of the time with our ears and brain? If that isn't feasable, what about a steerable directional microphone and listening to the strength of the sound?

My questions include what DSP would be appropriate (preferably with a low-cost development system)? How much off the shelf software is available and how much would I have to develop myself? What is the estimated magnitude of such a project?

Any advice, including partial advice, is greatly appreciated.

Bob

  • 0
    TI__Guru* 84110 points

    Bob,

    My recommendation is the TMS320C6748 for this project. There is a low-cost Experimenter Kit that will be a good development platform for you. Click on the Product Folder link in my first sentence and look for Tools.

    If you need more compute power and can work with fixed point numbers, then the TMS320C6424 would be a good choice. Or you can move to one of the newer C66x DSPs like the TMS320C6657, but I do not believe the EVM for the C6657 will have audio peripherals already on the board.

    You may find some useful information or loosely related examples on TI.com under Applications. Look through those to see if you can find some help or hints. Of course, a search on TI.com for the software items you are looking for would be a good idea, and then an Internet search for those would be the next choice.

    Regards,
    RandyP

  • 0
    Expert 2210 points

    If this was for a graduate college course, I'd say it should be a one-semester project.  You might consider latching onto a textbook based around one or more of the TMS320C6xxx DSP starter kits (DSK) and obtain the same DSK to facilitate the learning process.

    The field is often known as "acoustic location" (Google for information) and has been developed, for example, by the military in order to locate the origin of gunshots.

    For this kind of thing you really want transducers separated by a significant fraction of the relevant wavelengths.  Steering a single transducer is not at all the same thing.

    For audio signals, practically any TMS320C6xxx DSP should be fast enough for some of the simpler processing algorithms.  You need to consider how you're going to view the results - just an "event detected at 230 degrees +/- 15 degrees" or what?

  • 0 in reply to Douglas Gwyn
    Intellectual 355 points

    Thanks to all for your valuable information. I now know that it's more than I am ready to attempt. Darn.

  • 0 in reply to Bob Morris
    Expert 2655 points

    Grtns,

    I suggest you check the PADK from Lyrtech.  It is C6727 based and has all you need to do up to six beam-forming mic array.  They have good code example included with their product.

    http://lyrtechrd.com/en/products/view/+professional-audio-development-kit/

    Good Luck,