Hello,
I’m simply trying to create a ten second .wav 22khz tone and save it on my desktop. I’m a rank amateur at this stuff and a couple of hours of Googling brought me finally to Audacity. It appears to be perfect.
I"m sure it’s quite rudimentary. I wonder if someone wouldn’t mind taking through the steps required.
Thanks so much,
Frascati.
Sean Williams
You did pick one odd thing. The default sample rate of 44100 will not directly support 22000 tone.
Launch Audacity and Edit > Preferences > Quality. Change the first value, sample rate, to 96000 > OK.
Generate > Tone: Sine, 22000, 0.8 amplitude, and the duration you wish.
After it arrives, File > Export > WAV > Name it > Save.
Koz
We do note that scientists are routinely disappointed with Audacity because it’s an audio production editor not a WAV editor. Some technical shortcuts and inaccuracies are allowed in order to make your choral performance sound good.
Why would you want a 22000 sine wave (I’m assuming sine)? Your dog might be entertained, but very few humans are going to be able to hear it. Worse, if you play it in a sound system, you could fry the tweeter into smoking ruin. Your first indication of damage will be smell, not sound.
Koz
Are you aware that 22 kHz is inaudible and is too high a frequency for most audio equipment?
Why would you want a 22000 sine wave (I’m assuming sine)? Your dog might be entertained, but very few humans are going to be able to hear it. Worse, if you play it in a sound system, you could fry the tweeter into smoking ruin. Your first indication of damage will be smell, not sound.
Coincidence? Prescience? I’m actually trying to put together (i know, for the umpteenth unsuccessful time) an anti-barking file for a couple of the neighbors’ dogs. I did not mention that since it ends up being a distraction in any thread/conversation devolving into neighborly relations, animal cruelty, rights/responsibilities, etc, etc, etc.
I have a program called “Digital Dogsitter”. It’s ostensible purpose is to record your own soothing voice, or a soothing voice .wav file from the web, that will respond to your pet whenever it barks. If you are away at work or short vacation the software will run, voice/bark activated and calm your dog while you’re away.
I realized that the software could just as readily be used to package a desktop file of irritating tones to a horn loaded piezo directed at the fenceline so that when Eddie, the little terrier/terror mix next door is left alone by his owners to defend the property while they dine out till midnight, commences to snarling and barking at (what seems) 110db if I should as much as flinch in my own yard or driveway all evening long, ALL EVENING LONG, Digital Dogsitter would respond with a ten second blast of equally irritating noise to back him off.
I’ve read may tales of how ineffective these anti-bark tones are. I’m still determined to experiment with tones, warbles, sirens, white noise, or other, that is beyond the range of kids in the area, but irritating to dogs. Understanding that dogs hearing can range upwards of 45Khz and even the best young human ears cannot exceed 20Khz there seems to be ample room for experimentation.
Digital Dogsitter only accepts .wav files or microphone input. I’m using a five year old HP 8510w business class “workstation” with mostly best-of-class components installed for its time. I think the soundcard can handle this. I can supply the exact card id if necessary.
I noticed that Audacity has settings for “sound activated recording”. Does it handle sound activated playback? If so, I could dispense with the Digital Dogsitter entirely.
So the 22khz is not critical. Any tone that I can play with that exceeds 20khz and I can save and replay as a 10 second file would be ideal. If I can save it and load it into Digital Dogsitter it would be fine. It’s an extra simple GUI that allows threshold settings for voice level activation and play. If Audacity is capable of the same that is all the better. But right now it looks as if I need audacity to create a ten second .wav of 20k or above in order to save to Digital Dogsitter and run from there.
kozikowski,
I followed your directions and I can hear the 22Khz test tone quite clearly.
Something did not go right. I’m 52 and the limit of my upper range is nearer to 13Khz.
You may have 22KHz but you are listening to the distortion of your sound system not being able to handle it. If you could hear it, it would sound like a very quiet, delicate air-escaping hiss. We had a very tall, young engineer at a radio station that could detect the presence of 19.2KHz. It didn’t have to be loud, but he could do it every time. He couldn’t hear over that and nobody else can, either.
I always wondered what would happen if you recorded a very high quality track of his barking and played it back – in real time. Most dogs can’t keep up hysterical barking for a long time and it will eventually make him crazy.
You could also play a little bit of it before they leave. The dog will go nuts and they’ll be forced to deal with it. It doesn’t have to be loud, either. Dogs can detect other dogs walking quietly across the street through a fence.
I did find one way to successfully deal with this. Assuming you didn’t already alienate the owners. “Chucky,” a Chihuahua with a voice that could cut timber would protect the property across the street every time the owners left. I posted a note at the front door who I was with clear contact information and that there was a wild animal in the neighborhood (which there sometimes was) and I though Chucky may have seen it because of his barking. I wrote about how I thoroughly inspected their property, at midnight with a powerful flashlight, and as far as I could see there was nothing wrong. Chucky got moved to a different window – one that didn’t point to my house.
I also toyed with the idea of carefully recording the whole performance and delivering the whole thing to them on CDs. Most people have no idea at all the dog is doing this.
Koz
Oh, Audacity doesn’t do much in real time and Record plus Play is one of those. You might be able to jigger Overdubbing to do that… I don’t know if you can perform Overdubbing and Timed Recording at the same time.
Koz
What a strange idea - I had to try it 
Yes you can.
Disclaimer: doing this may not be a good idea. Use your own discretion.
What you need:
For this to work correctly there must be no passing aircraft / motor cycles / children / thunder storms, or any other sounds that could act as a sonic trigger.
Also note that high frequency sounds do not travel well and are easily blocked by objects (such as walls, windows, fences, trees… so ideally your playback transducer (loudspeaker) should be a highly directional horn device that is in open line of sight of the victim/dog.
The pick-up transducer (microphone) must be positioned so that it will pick up both the sound of the dog barking and the sound from the playback transducer (loudspeaker).
By the way, I’m using the term “transducer” because normal audio speakers and microphones roll off sharply in their responsiveness close to the upper end of the normal audio frequency range. and so will probably not be adequate for either playing, or picking up sounds over 20 kHz.
Adjust the sound activation level so that it is not activated by ambient noise - you only want this to “go off” when the dog barks.
Here we see Audacity set up ready to go.
What happens when the dog barks is that Audacity will record each bark, and will pause between barks.
When the cursor position reaches the start of the first sound in track 1, that sound will play (provided that “Transport > Overdub” is enabled). The sound from track 1 will be recorded on track 2, and because there is now continuous signal input, play/record will continue until it reaches the end of that first beep.
One quick note, you can’t stop and start tone groups like that without humans being able to hear ticks as each one starts and stops. AM/FM radio theory, modulation hoo-haa, etc. Fade each tone in and out should get rid of that. Effect > Fade-In, etc.
Koz
Wow, that’s some to chew on for a while. I’ll give it a go. I am first going to try the digital
dogsitter since it’s so dead simple. Just put a .wav file in a folder, dogsitter queues it up, set
the gui voice/bark response threshold (he’s pretty ear splitting so with a properly positioned mic
nothing short of a lawnmower or firecracker should false trigger it), and experiment with different
.wav samples online, bursts, sirens, white/brown/etc noise.
