Question: The Speed of Human Hearing

[Lo-Key Fu]

Hi All,

I'm guessing that some of you would already aware that 'perfect' human hearing usually ranges from frequencies of 20 Hertz to 20,000 Hertz; but I'm interested in the concept of how *fast* we can hear.

As in, how many microseconds difference a 'perfect' human ear can discern between two sound sources before the time between the sounds becomes too small?

Just curious really, but I'd love to know the answer.

Anyone?

[The Leveller]

Are you serious?

Sound has a fixed speed when it travels through air.

How fast your brain is able to determine between different sounds being detected by the ear is going to vary greatly from individual to individual based on a huge amount of variables.

But in all cases it could be said that the brain detects the vibration in the ear so fast it could be described as immediate.

[Kent_in_CO]

Awesome question. It makes sense that everyone would have their own unique "latency," even if it's really damn small.

[pepezabala]

well, you can try it out with your DAW. I suppose that it should be somewhere at 10 milliseconds. That's where you notice latency more or less. Everything below 10 milliseconds gets perceived as simultaneous event.

However there should be a wide range of variables coming in. Pitch, loudness, environment etc.

[Lo-Key Fu]

Are you serious?

Oddly enough, I am serious. Over tired perhaps, but serious nonetheless.



I'm aware that sound has a fixed speed of about 343 meters / second that is also affected by the medium it travels through, (ie. air, water); but I'm taking a punt and suggesting that body parts involved in hearing - like cillia - would hit physical limits of how fast they can actually respond (vibrate) before the neurons of most individuals.

As a clumsy example:

If we were to take two short percussive hits and place them a second apart, I doubt that the majority of people would have a problem with identifying them as two distinct sounds. While gradually moving those sounds closer together however - talking distances of microseconds if need be - at some point they would sound like a single hit to the human ear. I think this point would occur before the two sounds were dead level, but I'm entirely happy to be corrected if not.

So assuming the concept outlined above is actually correct, and in the context of this example, my question would be:
"Measured in time, what is the furthest distance apart these two sounds could be where they are still perceived by (average) human hearing as a single sound?"

Another awkward example?

Assuming a compressor reponds exactly at the attack time set on it's controls; below what setting would the difference of when the compressor starts to compress be imperceptable?

Or another?

When a DJ is beatmatching and gets a mix "in", assuming not every mix is scientifically perfect in terms of alignment, what would the error margin be for no perceptible difference for the audience?

It makes sense that everyone would have their own unique "latency," even if it's really damn small.

I tend to agree mate, and I'm very curious as to what the 'average latency' might be.

well, you can try it out with your DAW. I suppose that it should be somewhere at 10 milliseconds. That's where you notice latency more or less. Everything below 10 milliseconds gets perceived as simultaneous event. However there should be a wide range of variables coming in. Pitch, loudness, environment etc.

I was thinking it would have to be around that mark too, and I agree there would be all sorts of external factors involved (you mention some excellent ones). Not quite sure how I'd best go about testing this concept however, and I was hoping to find some info after a quick google earlier, but as yet, I haven't managed to turn up anything useful; hence my post.

[ark]

People can generally perceive two transients that occur about 20-30 milliseconds apart as separate events. Closer together than that and they merge.

If you take recorded speech and play it back twice, separated by about 50 milliseconds, it becomes much harder to understand.

Why do you ask?

[pbs99]

Check out the "Haas Effect", may answer some of your questions.

from wikipedia
"In 1951 Haas examined how the perception of speech is affected in the presence of a single, coherent sound reflection.[1] To create anechoic conditions, the experiment was carried out on the rooftop of a freestanding building. Another test was carried out in a room with a reverberation time of 1.6 ms. The test signal (recorded speech) was emitted from two similar loudspeakers at locations 45° to the left and to the right in 3 m distance to the listener.
Haas found that humans localize sound sources in the direction of the first arriving sound despite the presence of a single reflection from a different direction. A single auditory event is perceived. A reflection arriving later than 1 ms after the direct sound increases the perceived level and spaciousness (more precisely the perceived width of the sound source). A single reflection arriving within 5 to 30 ms can be up to 10 dB louder than the direct sound without being perceived as a secondary auditory event (echo). This time span varies with the reflection level. If the direct sound is coming from the same direction the listener is facing, the reflection's direction has no significant effect on the results. A reflection with attenuated higher frequencies expands the time span echo suppression is active. Increased room reverberation time also expands the time span of echo suppression.[2]"

[photonal]

Would the frequency of the sounds also play a role in how close sounds would be perceived?

Quote : "The range of human hearing is generally considered to be 20 Hz to 20 kHz, but it is far more sensitive to sounds between 1 kHz and 4 kHz. For example, listeners can detect sounds as low as 0 dB SPL at 3 kHz, but require 40 dB SPL at 100 hertz (an amplitude increase of 100). Listeners can tell that two tones are different if their frequencies differ by more than about 0.3% at 3 kHz. This increases to 3% at 100 hertz. For comparison, adjacent keys on a piano differ by about 6% in frequency." ( http://www.dspguide.com/ch22/1.htm )

[The Leveller]

The variables are so huge it's not worth worrying about if you ain't got the specifics. What type of sound/wave? Are they the same pitch or different pitch? what volume? how old is the person? etc.etc.

but I guess if you mean from a production point of view then the boys above have got yer answers.

[Lo-Key Fu]

People can generally perceive two transients that occur about 20-30 milliseconds apart as separate events. Closer together than that and they merge. If you take recorded speech and play it back twice, separated by about 50 milliseconds, it becomes much harder to understand.

Why do you ask?

Thanks very much for this info! That's exactly the kind of answer I was hunting for. May I ask how you came across these figures? I'm not doubting your information whatsoever, just keen to read and learn more.

Why?

Out of sheer curiousity if truth be told; but I was at the tail end of a number of hours working away on some tunes, and caught myself adjustinga number of time based effects settings a little too intricately. I decided to take a break, and the thought crossed my mind that at some point I would no longer be making any tangible difference to the end result; so I spent my down time attempting to put some sort of quantifiable value on it.

Well, you did ask...

Check out the "Haas Effect", may answer some of your questions.

Excellent and thanks! I'm heading over there now for a bit more of a read. Very much appreciated mate.

Would the frequency of the sounds also play a role in how close sounds would be perceived?

My first response would be "Yes, I think so". And volume would definitely be a factor too. Anyone ever recorded an instrumentalist in the studio and by varying only their headphone volume you can actually adjust whether they're playing before, on or after the pre-recorded beat or click?

[Lo-Key Fu]

The variables are so huge it's not worth worrying about if you ain't got the specifics. What type of sound/wave? Are they the same pitch or different pitch? what volume? how old is the person? etc.etc.

but I guess if you mean from a production point of view then the boys above have got yer answers.

Fair call mate, and I agree with your perspective, but it was just something that piqued my interest (for reasons mentioned above), and I figured there would have to be some kind of answer, even if it was going to be drastically affected by other factors.

That said, the more we understand about our medium the better in my book.

[ark]

Thanks very much for this info! That's exactly the kind of answer I was hunting for. May I ask how you came across these figures? I'm not doubting your information whatsoever, just keen to read and learn more.

I used to work with a guy who did acoustics research for a living. In fact, he's one of the people who did the original research for MP3 encoding. He also did theatrical sound reinforcement.

Once upon a time I mentioned to him that I had a hard time understanding dialog in a particular theater. My wife and I had what we thought were excellent seats, and we could hear everything on stage clearly; we just couldn't understand what people were saying.

As it happened, he had done sound reinforcement work at that particular theater, so he explained to me about the problems that come about when you hear sound directly from the stage, and also sound reflecting off the (hard) ceiling, which therefore takes a longer path to reach one's ears.

More recently, I was talking to another friend who was working on sound reinforcement in a completely differen concert venue with a similar problem. In this case, the stage had a reflective rear wall, and the sound from the stage monitors was reflecting off that wall and reaching the audience's ears about 30 milliseconds after the sound from the mains. I suggested to him that the delay might be the problem, and he said "Of course!" and got his hands on a delay unit that would make the sound from the mains come out 30 ms later. It made an amazing difference in how easily the audience could understand lyrics.

So I guess the short answer is "It's just one of those things I picked up from people who do this kind of stuff for a living."

[crumhorn]

It's interesting that 20 to 30 events per second seems to be the point where we loose the ability to separate them. e.g. a series of clicks becomes a tone or a series of still images becomes a movie.

It's also applied in human interface design. If you press a button and a sound goes click within 20 to 30 ms then you will perceive the sound as directly connected to the button.

I remember being told it's called the Human Cycle Time but I can't find any exact matches on the internet - although It did turn up this article which seems relevant --> http://en.wikipedia.org/wiki/Human_info ... ssor_model

[Angstrom]

10 ms is the allowable latency for hearing-aids, anything longer than that starts to be discernable as separate from the originating event, less than 10ms is perceived as part of the event.


As a guide to human latency: An electrical pain applied to the human wrist is detected in the brain 3ms later. Obviously ears are closer to the brain than the wrist is.

It's still safe to say that we perceive the world as it was, not as it is.

[Lo-Key Fu]

10 ms is the allowable latency for hearing-aids, anything longer than that starts to be discernable as separate from the originating event, less than 10ms is perceived as part of the event. As a guide to human latency: An electrical pain applied to the human wrist is detected in the brain 3ms later. Obviously ears are closer to the brain than the wrist is.

It's still safe to say that we perceive the world as it was, not as it is.

Brilliant Angstrom, and thanks!