How can a black hole produce sound? I was reading this article from NASA -- it's NASA -- and literally found myself perplexed. The article describes the discovery that black holes emit a "note" that has physical ramifications on the detritus around it.

Sept. 9, 2003: Astronomers using NASA’s Chandra X-ray Observatory have found, for the first time, sound waves from a supermassive black hole. The “note” is the deepest ever detected from any object in our Universe. The tremendous amounts of energy carried by these sound waves may solve a longstanding problem in astrophysics.
The black hole resides in the Perseus cluster of galaxies located 250 million light years from Earth. In 2002, astronomers obtained a deep Chandra observation that shows ripples in the gas filling the cluster. These ripples are evidence for sound waves that have traveled hundreds of thousands of light years away from the cluster’s central black hole.
“The Perseus sound waves are much more than just an interesting form of black hole acoustics,” says Steve Allen, of the Institute of Astronomy and a co-investigator in the research. “These sound waves may be the key in figuring out how galaxy clusters, the largest structures in the Universe, grow.”

Except:

*

*Black holes are so massive that light, which is faster than sound, can't escape.

*Sound can't travel in space (space has too much, well, space)

*It's a b-flat?

So: How can a black hole produce sound if light can't escape it?
 A: Actually, its not the Black Hole (singularity or anything below event horizon) which is emitting sound waves.
A Black Hole sucks matter from its neighborhood.  These being sucked matters produce sound waves when they are on its way before entering Event Horizon of black hole (light can't escape from inside event horizon). The produced sound waves are nothing but cavity-based ripples in Perseus cluster gas. The ripples are formed by "jet of materials" (traveling at near speed of light) pushing back cluster gas. Due to its high energy, it can be detected at millions of light years from source.
But, what expels matter at such high speeds near the ultimate sinkhole?
The answer: Its the enormous electromagnetic forces which shot it away before it got beyond the black hole's event horizon. These powerful forces are generated as magnetized, hot gas swirls toward the black hole creating extreme voltages which accelerate particles away from the disk in opposite directions.
A: I'm not going to address the production mechanism,1 just the nature of the "sound" in this case.

What you think of as the hard vacuum of outer space could just as well be seen as a very, very, very diffuse, somewhat ionized gas. That gas can support sound waves as long as the wavelength is considerably longer than the mean free path of the atoms on the gas.
As for the tone, there is a simple relationship between the tone of the same name in different octaves, so once they know the dominant frequency they can figure its place on the scale.

1 Though it won't be happening inside the event horizon -- which is where "not even light can escape" holds -- but in the region around the hole proper where it accumulates gas and dust and the magnetic fields from the hole play merry havoc with the ionized components of the accumulated stuff.
A: As the other guys have already covered most of the topic, I'd like to quote some things. Light can't escape only from the inside of event horizon because it has already fallen into it. But after reading the article now, we could indicate some points.


*

*The article specifically says a "supermassive blackhole". They're a way too bulk in size when compared to other blackholes (Schwarzchild for example). So, they're not strong enough to produce an astonishing tidal force upon the falling matter at the event horizon. And that's because the central singularity is located very deep inside the BH (quite a large distance from the event horizon).

*As you say, outer space is just hard kinda vacuum and not completely vacuum. It consists about a few hydrogen atoms per cubic meter. In your case, the gases in the Perseus cluster is good enough to serve as a medium for these very low frequency sound waves to travel through.
The sound waves are just longitudinal pressure waves and hence the fate "they require medium". As the BH pulls material inwards (into it), the gravitational pressure causes it to expel matter and energy out of it which we perceive as gas jets and we named them as "relativistic jets".

Astronomer Steve Allen told, (I don't know him, but looks like he's a big guy)

The ripples were caused by the rhythmic squeezing and heating by the intense gravitational pressure of the jumble of galaxies packed together in the cluster. As the black hole pulls material in, it also creates jets of material shooting out above and below it, and it is these powerful jets that create the pressure that creates the sound waves.

But I'm happy that we haven't really heard the black hole sound. The actual observation was due to X-Rays. I can't even imagine, "How sound waves reach us traveling through the bare leftover hydrogen atoms distributed throughout the galaxy?" Indeed, they may be able to travel through the jets but can't reach us that fast..! Actually, these pressure waves have traveled along with the jets traversing throughout the medium by their stretch/squeeze physical fantasy. I think the long-term curiosity got vanished by this kind of observation.   
Why these X-Ray clouds are always hot and don't cool down at all?
The same topic spoken around NASA indicates something satisfiable...

The sound waves were indirectly detected using the Chandra telescope because the cluster gas is very hot and thus emits an especially energetic form of light called X rays, as well as less energetic visible light. And the gas is so hot because of the effects of the black hole. More than an acoustic curiosity, these sound waves transport energy that keeps gas throughout the cluster warmer than it would otherwise be. These warmer temperatures, in turn, regulate the rate of new star formation, and hence the evolution of galaxies and galaxy clusters.

And your article too agrees the fact:

The sound waves, seen spreading out from the cavities in the recent Chandra observation, could provide this heating mechanism.

A: First, for additional references, there is the original press release. Also, a similar report from a different black hole is here. It seems the Chandra people like this sort of thing. It is also worth noting that as far as I can tell, there are only press releases and no published scientific articles on this phenomenon.

Now to address the questions.

Black holes are so massive that light, which is faster than sound, can't escape.

Well, light from inside the black hole cannot escape. But active black holes create violent neighborhoods around them. In general, there will be an accretion disk - a relatively flat disk of material slowly spiraling into the black hole. Friction due to differential rotation in this disk can make it glowing hot, especially close to the black hole. Furthermore, angular momentum and magnetic fields conspire to make jets shooting out of the "poles" of the black hole (pretty much all black holes in astronomy are expected to be rotating significantly). Again, the matter never actually got into the event horizon, since by definition we would not see it again.
So what is this sound? As best I can tell from the press releases, these particular black holes cycle through periods of low accretion and high accretion. There may be a lot of material falling in for a long time, powering jets that pump energy out beyond even the galaxy in which the black hole resides.1 Then there will be a few million years of almost nothing falling into the black hole, during which the jets are little more than trickles. This cycle repeats semi-regularly, causing periodic bursts of energy to be sent out.

Sound can't travel in space (space has too much, well, space).

Well, yes and no. It is true there is some material even in the space between galaxies. On the other hand, it is extremely diffuse. In fact, it has been said (I cannot remember the source) that the densest clouds of material in interstellar space are more diffuse than the best vacuums we can make in laboratories. So you might imagine there can be some propagation, but not in the traditional sense.
Really, when the Chandra people say there is "sound," what they mean is this. Thus pulses of energy sent out form shock waves2 that propagate through the intergalactic medium. If you look far enough, you will see these periodic dense regions in a pattern not unlike the ripples in a pond. Since "sound" in the normal sense consists of overdensities traveling through the air (albeit without shocks in most cases), and since these are periodic overdensities in the diffuse gas between galaxies, we may as well make a connection between them in terminology.

It's a b-flat?

Take this with a large grain of salt. I doubt the duty cycle of the black hole is so regular as to produce a monochromatic "pitch." This is more of a whimsical calculation. Nathaniel in a comment did the reverse calculation - going from "note" to frequency. To see how the scientists changed a frequency $f$ (basically the reciprocal of the time between active periods of the black hole) to a named note, see this Wikipedia entry. In short, the number of the pitch is
$$ p = 69 + 12 \log_2\left(\frac{f}{440~\mathrm{Hz}}\right). $$
They plugged in an $f$ and must have gotten a $p$ around $-614$ or $-626$ (the wording is ambiguous). The B-flat above middle C has $p = 70$, the one below middle C has $p = 58$, the next one down is $p = 46$, etc.

Addendum: Why is this interesting? Beyond the whimsy, there is scientific value to this. Believe it or not, most of the mass (or at least, most of the mass of normal, non-dark matter) of a cluster of galaxies is found outside the galaxies themselves, in this intergalactic (aka intracluster) medium. It makes up a significant portion of the universe despite its low density, due entirely to the large volume it occupies. More interestingly, this gas is very hot - millions of degrees, despite the "background temperature" of the universe arguably being a few degrees above absolute zero. It is so hot that it glows in X-rays, which is why Chandra - a space-based X-ray telescope - studies it. These sound waves might explain a significant source of heating and injection of turbulence into this matter, which can have broad implications for extragalactic cosmology and galaxy formation. This all falls under the key term "AGN feedback," which is quite a hot topic in astrophysics these days.

1 Our own Milky Way's supermassive black hole probably had an active phase at some point. Astronomers recently detected the so-called Fermi bubbles that indicate two jets were shooting out of the center of our galaxy millions of years ago.
2 "Shock waves" have a technical meaning in fluid dynamics. Basically, there is a discontinuity in the density - ahead of the shock, everything is normal and diffuse, but as it passes there is almost instantaneous compression and heating. Think of the blast wave from an explosion.
A: I am going to speculate on a production mechanism to complement @dmckee's answer.
It is true that light cannot escape from a black hole, except it can lose energy through Hawking radiation.
Suppose the black hole oscillates,vibrates,  i.e. within it compression waves exist this would of course be another way of losing energy. This because the radius of the horizon would be changing in time, i.e. the gravitational field in space around it.
The detection of these sound waves around the hole would also be a measurement of  gravitational waves .
I googled "vibrating black hole" . the first hit:

A black hole can vibrate, and its vibrations produce gravitational waves (ripples in the fabric of spacetime). These waves carry away the "hair" of any newborn black hole, leaving it hairless when quiescent, and they also carry encoded in themselves the values of the hole's mass and spin [Bill Press, Richard Price, and Saul Teukolsky].

A second way of gravitational waves from the same source:

A small black hole orbiting a massive hole or any other massive body produces gravitational waves, and those waves carry, encoded in themselves, full maps of the body's warped spacetime. If we can detect the waves and extract their maps, we can use the maps to determine the nature of the massive body, and if it is a black hole, we can use the maps to test the no-hair prediction [Kip Thorne and Fintan Ryan].

So it is a scientific speculation and not a science fiction one.
A: 
So: How can a black hole produce sound if light can't escape it?

Light can't escape a black hole only if it's inside the event horizon. Light can escape (the gravitational pull of) a black hole if it's outside of the horizon. Likewise, sound can escape the gravitational pull if it's sufficiently far away from the horizon. I would guess that the horizon is different for light and for sound, but that's just a hunch.
A: TL;DR

Basically the black hole pulls in particles BEFORE the event horizon,
  those location where those particles were is now void, "Nature abhors
  a vacuum" so that void is filled with nearby particles, those
  particles now leave a void, so that void is filled with further out
  near-by particles, and so on, and so on, creating a ripple, a wave, a
  sound. Granted it will be very low frequency and inverse to a
  traditional sound wave since the particle are moving towards the
  source rather than away from it.


I'd be more concerned about how they are detecting this sound than how
it escapes... I'm not aware of any medians in space for sound to
travel through but then again I've never studied space... or physics
for that matter..
I do however know that you can thing of sounds as compression and decompression of particles in an array... When anything gets pulled away it leaves an empty space that will get filled by particles around this space (vacuum effect... pressure pushes these particles in) when those particles leave they now have created empty spaces, so now particles around them (probably even including themselves) go back toward that space.... and so on and so one, you create a ripple effect (keep in mind this is a 3D ripple effect... picture wobbling spheres inside one another, not a stone hitting a body of water haha) This ripple (3D ripple) can be considered a wave, it has a frequency, and therefor when this motion of compression reaches an eardrum it will be detected...
Again, I still don't know how this sound travels anywhere since space is in itself a vacuum from what I know...
