OK, now to provide feedback / answers to some of your comments and questions:
- Since red foxes hear in the ultrasonic range (up to around 50kHz), I would need a sampling rate of 96kHz so that the Nyquist value represents the fox’s full hearing range.
Agreed. Tympan does up to 96 kHz, as you said, which is hopefully good enough.
- Because people cannot hear into the ultrasonic range, I would need real-time frequency compression (dividing input frequencies by 0.5) so that ultrasonic details become audible.
Agreed, though remember that most living humans (especially those over 25 years old) can really only hear up to ~16 kHz, and even that can be difficult. So, yes, you will need signal processing to bring the ultrasonic audio down to the human-audible range, but you might need to try a couple different approaches and find which is most satisfying for you and your demo:
- Compress the whole frequency range of 0-48 kHz down to 0-24 kHz (or 0-16 kHz or whatever). While this sounds easy, the signal processing algorithms often introduce weird artifacts that can be very distracting.
- Or, you can leave the audible frequencies untouched and then you can shift just the ultrasonic frequencies downward so that they are audible. You’d be hearing the natural audible audio mixed with the artificially down-shifted ultrasonic audio. While this seems less ideal than the compression approach discussed in (1), the fact that the audible frequencies are left untouched means that the listening experience becomes immersive more quickly.
There is a Tympan example for frequency compression and for frequency shifting, so you can build from working code and then quickly see which you prefer.
Be aware that the Tympan has small two microphones that are built-into the circuit board. These built-in mics do work for ultrasonic frequencies! So, without building anything yourself, you can use the Tympan plus its microphones to listen in the ultrasound. You can quickly decide which signal processing approach will work best for you.
- An audio receiver would be placed within 3D-printed fox pinnae (external ear structure), which would be worn over the head; I would need a receiver for each ear.
- Input audio would be transmitted to closed-back headphones (potentially via Bluetooth). The ears would be attached to the headband part of the headphones.
Agreed. As shown earlier, we did the same thing with human ear shapes and we put a microphone in each ear canal. We put speakers in our earcups, just like you did. Our BOM is part of the repo that I linked earlier (including 3D printing costs). It’s an old project, but it still might be helpful.
- Are there student discounts available for the Rev F? My worry is that, if I purchase the Rev F, I will be required to buy further components that I cannot afford. Even if there are no discounts, would it still be worth purchasing the Rev F for this project?
Sadly, I can’t help you with this question. @biomurph might be able to let you know. As for whether the Rev F is “worth” it, I can’t answer that question for you. I can tell you:
- The Tympan hardware/software was indeed designed to manipulate audio, like what you want to do. It seems like a good choice to me.
- But, while there are example programs that are pretty close to what you want, you will likely need to be able to do some programming in Arduino (which is a flavor of C++). Are you comfortable with doing a bit of programming?
- Is it possible to use Tympan as a modified hearing amplifier, or are there parts of my project that will not work with Tympan?
Thinking about your project, if you want to put microphones in your fox-shaped ears, you will need:
- Your 3D printed fox ears
- Microphones (electret, a few dollars)
- Wires (a few dollars)
- A plug to connect to plug the wires into the Tympan (a few dollars)
- Some wired heaphones or earbuds (hopefully you might already have some?). Sadly, Tympan won’t do a bluetooth audio connection, sorry.
Unless you have very strange microphones, the Tympan has a modest mic pre-amp that can handle most common low-cost mics, such as electret mics. The Tympan also has a modest headphone amp to drive most consumer headphones/earphones (down to 16 ohms). So, you shouldn’t need anything extra for that.
- Since I’m new to programming, what are some resources for DSP coding and working with the Tympan circuitry? Is there a particular area of programming I should start with?
This is the biggest issue with any of these hobby audio projects. Do you have any previous programming experience?
We targeted users who had some experience using Arduino electronics and Arduino programming. Have you done this before? If so, then you’ll feel at home and you’ll be able to focus on the DSP. If not, you’ll have to learn to do a bit of programming, too, which will be effort.
Let me know how all of this sounds!
