Dog whistle a high training whistle in your browser

How to use the online dog whistle

Choose a pitch, choose a pattern, start quietly and raise the volume only as far as you need. The whistle on this page is the site's tone generator limited to 8,000–22,000 Hz, with four whistle patterns: a continuous tone, beeps, short bursts and a double whistle.

The pitch you pick decides who hears it. Around 8–12 kHz almost everyone hears the whistle clearly, including you, which is useful when you want to be sure it is playing. Around 16–20 kHz many adults hear little or nothing while dogs still hear it easily. The top of the slider depends heavily on your speaker: small phone and laptop speakers and most headphones fall off steeply above about 18–20 kHz, so a very high setting may simply come out too weak to matter.

The pattern matters as much as the pitch. Short repeated pips and a single long blast are the two signals trainers use most, for reasons explained in the training section below. Pick one pattern for one meaning and keep it that way.

• Start at a low volume with the speaker pointed away from the dog, never held close to its ears.
• If you cannot hear the tone yourself, do not turn it to maximum: the dog, children nearby and your speaker may be getting far more sound than you think.
• Keep each whistle short. A signal of a second or two is enough; long continuous high tones are unpleasant for animals and hard on small speakers.
• Stop at once if the dog flinches, cowers, paws at its ears or leaves the room.

What frequency is a dog whistle?

Most dog whistles sound between 23 and 54 kHz, which is above human hearing, and some are adjustable down into the audible range. To a person, a so-called silent whistle usually gives only a faint hiss of air. That figure comes from Wikipedia's article on the dog whistle, which cites Coile and Bonham (2008), so treat it as a typical range across products, not a standard.

For comparison, the textbook range of human hearing is 20 Hz to 20 kHz, and that describes young ears. Anything above about 20 kHz is called ultrasound. A dog hears up to about 45 kHz, so a whistle in the 23–45 kHz band is in the zone that dogs hear and people do not. That gap is the whole idea of the silent whistle.

An audible high whistle, like the one on this page, works on the same principle with a smaller gap. At 18–20 kHz most adults over 50 hear nothing, younger people may hear a thin whine, and a dog hears a clear tone close to its most sensitive range.

Can a phone or browser play a silent dog whistle?

No. A phone, a laptop or a web page cannot play the 23–54 kHz of a real silent whistle, and any app that says otherwise is overstating what the hardware can do. The reason is a hard rule of digital audio: a digital signal can only carry frequencies below half its sample rate, the Nyquist limit.

The two common sample rates are 44.1 kHz, the CD standard, and 48 kHz. Half of those is 22.05 kHz and 24 kHz. In a browser, the Web Audio API runs at the sample rate of its audio context, normally 44.1 or 48 kHz, so the ceiling is about 22 kHz. That is why this whistle stops at 22,000 Hz. On top of that limit, small speakers and most headphones lose output steeply near the very top of the range.

What a browser whistle can do is still useful. It plays high tones up to about 20–22 kHz that dogs hear well and many adults barely hear, and it plays clear, identical whistle patterns every time. For training, the repeatable signal is the part that counts.

What frequencies can dogs hear?

Dogs hear from about 60 Hz up to about 45 kHz, more than twice as high as young people. In Henry Heffner's tests of four dogs weighing 4.5 to 30 kg, the dogs heard tones at 60 dB SPL from 62 Hz up to 45 kHz, with their best sensitivity at 8 kHz. A Louisiana State University summary by George Strain puts the dog's best hearing at about 4–10 kHz and notes that the most sensitive dog detected a 4 kHz tone at −4 dB.

Size makes no difference to how high a dog hears. Heffner's 1983 study compared a poodle, a dachshund, a Saint Bernard and a chihuahua and found no relation between head size or eardrum size and high-frequency hearing. A chihuahua does not out-hear a Saint Bernard.

Hearing ranges always depend on how loud the test tone is, so figures from different studies do not line up exactly. The table below gives approximate ranges from the LSU comparison, which applies one cut-off to all species, next to single-species studies with their own criteria. The LSU page itself warns that the underlying methods differed, so small differences between species should not be over-read.

Two patterns stand out. Small mammals hear very high: Heffner and Masterton concluded that small heads go with high-frequency hearing, and that in land mammals good high-frequency hearing seems to come at the cost of low-frequency hearing, which is why a mouse hears nothing below about 2.3 kHz. Pet birds sit at the other end: by the LSU figures a canary or budgerigar tops out around 8 kHz, at the bottom of this whistle's range.

Can humans hear a dog whistle?

Humans cannot hear a true ultrasonic whistle above about 20 kHz, but many people hear the lower part of the dog whistle range, and young people hear much more of it than adults. In a study of 162 adults with normal standard hearing (Wang and colleagues, 2021), everyone under 30 heard 16 kHz and about half heard 20 kHz. No one aged 51–60 heard 20 kHz, and no one aged 61–70 heard even 18 kHz.

This is exactly what Galton noticed in the 1880s, when he wrote that older people were "absolutely deaf to shrill notes which the younger persons hear acutely". So if two people stand next to the same speaker playing 18 kHz, a teenager may wince while a grandparent hears nothing at all. Neither of them is wrong; high-frequency hearing fades with age, and that is normal.

Some devices sold as silent are not silent to everyone. When van Wieringen and Glorieux (2018) tested an ultrasonic rodent repeller on 25 young and 25 middle-aged adults, the device also gave off a faint audible sound around 4–5 kHz, and the younger people often heard its lower settings, 12–14 kHz and about 25 kHz. Several found them disturbing. The hearing test on this site lets you check where your own upper limit lies.

Who invented the dog whistle? Francis Galton

The dog whistle began as a hearing test for people. Francis Galton made his whistle around 1876, according to a 2009 history in the APS Observer, and described it in his 1883 book Inquiries into Human Faculty and Its Development. In his words, he "contrived a small whistle for conveniently ascertaining the upper limits of audible sound in different persons". It was a brass tube with a sliding plug marked for exact pitches, and Galton put the normal upper limit of human hearing at about 18 kHz.

Then he turned it on animals. Galton walked through the London Zoological Gardens with the whistle fixed to the end of a walking stick and a rubber bulb under the handle, squeezed it near the cages and watched whether the animals pricked up their ears. His verdict: "Of all creatures, I have found none superior to cats in the power of hearing shrill sounds." Insects were a dead end: "My attempts on insect hearing have been failures."

In town streets, he wrote, the whistle "made nearly all the little dogs turn round, but not the large ones". A century later Heffner's measurements found no real difference between small and large dogs, so Galton's street observation was probably about attention or temperament, not hearing. Galton whistles went on to be used to test hearing in rats, bats, insects and other animals.

How to use a dog whistle for training and recall

Use several short, quick pips to call a dog to you and one long, steady blast to slow it down or stop it. That pattern has research behind it. In a 1990 study, Patricia McConnell trained young dogs to come to one signal and to sit and stay to another. Four short rising notes were more effective at making the dogs come, and raised their activity more, than one long note did. One long descending note was not better for sit and stay.

McConnell then looked at what professional trainers actually do. In a 1991 chapter she reported that animal trainers working in many different languages use short, rapidly repeated notes to get animals moving and long continuous notes to slow them down. The lab result and centuries of handlers' practice point the same way.

Trainers give two more reasons for preferring a whistle to the voice: the sound is the same whoever blows it and whatever mood they are in, and it carries further than a shout. No study has compared whistle recall and voice recall directly, so these are practical reasons, not measured ones.

• Keep one meaning per pattern: for example short pips for come, one long tone for stop. The double-whistle pattern can be a third, separate cue.
• A whistle has no built-in meaning for a dog. It learns what the sound means from what reliably follows it, so start at close range in a quiet place before trying it outdoors.
• Use the same pitch every time. The advantage of a whistle is its consistency, and changing the frequency between sessions throws that away.
• Pick a pitch you can hear yourself, at least at first, so you know exactly what the dog heard.

Cat whistle: can cats hear high-pitched sounds?

Cats hear higher than dogs, so every tone this whistle can play is well within a cat's hearing. Rickye and Henry Heffner measured the domestic cat's range in 1985: at 70 dB SPL cats heard from 48 Hz to 85 kHz, which the authors called "one of the broadest hearing ranges among mammals". Cats gained their high-frequency hearing without losing the low end. The LSU table gives 45–64 kHz because it uses a different cut-off; both say the same thing in practice.

Galton reached the same conclusion with nothing but a whistle on a stick: no creature he tried beat the cat at hearing shrill sounds. A high tone that an adult owner cannot hear may be loud and clear to the cat on the sofa.

A cat whistle works the way a dog whistle does: as a signal the animal learns to link with something, such as food. Hearing the tone is not the same as obeying it. Because cats hear these frequencies so well, keep the volume modest and the tones short, and do not use high tones to startle or chase a cat.

Do ultrasonic rat repellent sounds and mosquito repellers work?

Controlled tests have found that ultrasonic pest repellers do not give useful protection, and a phone or browser cannot even produce ultrasound. For mosquitoes the evidence is clearest. A Cochrane review (Enayati, Hemingway and Garner, 2007) pooled ten field studies of electronic mosquito repellers: all ten found no difference in the number of mosquitoes landing on people with or without the device. The authors concluded that the devices "have no effect on preventing mosquito bites". No published test of a mosquito-repellent phone app was found; the apps imitate the devices that failed.

Other pests fare no better. In a test of four commercial ultrasonic devices, female bed bugs were just as likely to be found with the sound as without it (Yturralde and Hofstetter, 2012). In an Australian test of nine devices against paralysis ticks, the best moved 19.5 % of ticks away, which the authors called "insufficient to offer adequate protection" (Panthawong and colleagues, 2021).

For mice and rats the problem is habituation. The University of California's pest management programme notes that mice "quickly become accustomed to regularly repeated sounds", that ultrasound is directional, does not reach behind objects and weakens quickly with distance, and that "there is little evidence that sound, or magnetic or vibrational devices of any kind will drive established house mice or rats from buildings". In 2001 the US Federal Trade Commission sent warning letters to more than 60 makers and sellers of ultrasonic pest repellers, telling them their claims needed scientific proof; between 1985 and 1997 it had already taken action against six companies over such claims.

So a "rat repellent sound" played from a phone has two problems at once: the device cannot reach the frequencies the products claim to use, and the products themselves have not worked in tests.

Is a high pitch dog whistle sound safe for dogs, pets and people?

A short whistle at moderate volume is a normal training signal; the risk comes from loud, long or close-up high tones. No study has set safe levels for dogs listening to a speaker, so the rules here are practical ones: keep the volume moderate, never hold the speaker near an animal's ears, do not play long high tones near pets, and stop if the animal shows distress.

Remember the other animals in the house. Hamsters, rats, mice and guinea pigs hear these frequencies too, often better than dogs, and they cannot leave the room. Do not play the whistle next to their cages.

People need the same care. The WHO–ITU safe-listening standard sets the weekly limit for adults at 80 dB for 40 hours and for children at 75 dB, and the WHO advises keeping device volume at no more than 60 % of maximum. High tones can be uncomfortable long before they seem loud, and an adult who cannot hear a tone may turn it up without noticing what a child nearby is hearing. The Council of Europe made the same point about the "Mosquito" anti-loitering device, a 16–18.5 kHz tone: babies and children reacted strongly while their parents heard nothing.

Speakers have limits as well. Small speakers and tweeters are not built for long, loud, pure high tones, and a continuous loud high-frequency sine wave can overheat a tweeter. Start quiet, raise slowly and keep each whistle short.