Sibilance. A Good Thing?

[BEC]

OK, if I understand what you are calling sibilance, I consider it a good thing for letting you know your amp, tweeters and crossovers are working right. It is on some recordings and is in the higher regions of sound reproduction. I have a recording of Jerry Keller spitting out the song "Here Comes Summer" that I often play to test the highs out of tweeters. If your system is working right, this should be hard to listen to. You should be dodging spit from the tweeter and wishing it were over. But, it is there, it is recorded. If you don't hear it and hear it loud and obnoxiously clearly, something is wrong with the way your system reproduces highs.

Download this and get ready for a happening.

www.bandksound.com/sib.mp3

Bob Crites

[DrWho]

However, this test won't let you know if your system overexagerates the sibilance frequencies. I don't have any system that exagerates it so I can't think of a song off the top of my head that would reveal such a system, but you should have two different songs...one that indicates the HF capabilities of your system and one that indicates that it isn't overexagerated/distorted (a distorted high end also exagerates sibilance...prob has something to do with the similarity between the harmonics)

[Duke Spinner]

Man ....

what freakin mike did they use ...???

tweren't no Neunmann .....

[boom3]

I always understood sibilance to mean an extra hiss or "tail" on treble percussion, or perhaps at the very top end of a alto's voice. It was once explained to me as beats between harmonic distortion peaks, indicating that the tweeter diaphram is in severe breakup at some combination of power input and program content. As distinct from mere brightness, which is a uniform exaggeration across the treble range. I think that with Heritage products I am now spoiled to hearing horn treble. Just today I was listening to some Phillip Glass and the triangle and chimes were simply there, along with the rest of the music of course. I have noticed about my CW IIs that I can hear the persistant ringing within a treble percussive instrument, after it is struck, be it triangles, bells, chimes or cymbals. I can't hear that as well with direct-radiator tweeters, possibly because of the hysterisis of their much-heavier diaphrams.

[BEC]

----------------

On 8/20/2005 4:37:20 PM DrWho wrote:

However, this test won't let you know if your system overexagerates the sibilance frequencies. I don't have any system that exagerates it so I can't think of a song off the top of my head that would reveal such a system, but you should have two different songs...one that indicates the HF capabilities of your system and one that indicates that it isn't overexagerated/distorted (a distorted high end also exagerates sibilance...prob has something to do with the similarity between the harmonics)

----------------

Right, but what I am saying is that the sibilance is on this recording. Trying to do something about it on your equipment is detrimental to the sound of your other recordings. I can completely get rid of all of it from this particular recording by just turning off the tweeter. None of it is in the range of the K-55V. And yes, I have a total of 7 recordings I regularly use to evaluate how the top end is doing. They are not necessarily recordings I like, just that I am very familiar with them and after repair of a tweeter, they let me know pretty quickly if I got it right.

Bob

[DrWho]

well i was just trying to say that a speaker that can reproduce that nasty sibilance doesn't neceasarily mean that the speaker in perfect working order (ie, it could add sibilance to material that doesn't have any). But ya, it's a great way to see if that tweeter is at least working or not.

[D-MAN]

I find Diana Krall to have a problem with recording sibilance to the point of annoyance. Bad microphony, IME.

DM

[Erukian]

I've noticed that mainly Klipsch speaker's suffer from sibilance.

Let me explain, this can be both a good and bad blessing. On good recordings, the extra detail sounds awesome. On some recordings, comparing a Klipsch horn loaded tweeter to a soft dome will make you honestly want the soft dome. This is especially the case with some of Diana Krall's recordings.. They just have a slight edge that just grinds on you after a few songs and you want to stop listening to the disc.

At least this is from my experience.

-Joe

[dragonfyr]

Bob, you are correct in identifying sibilance as an undesireable artifact of the recording process.

It is unfortunate, as there is no reason for it to intrude into the recording if properly engineered.

Although once it has, it cannot be selectively removed without removing all of the program material that occupies the same frequency niche.

Thus you can appreciate a little more of it takes to be a good engineer and what the technical awards are about for production. And there is no reason for it to ever be present in a good recording.

Here is an article that addresses some of the factors and ways to correct it - to 'de-ess' the recording process that you might find interesting...

http://emusician.com/mag/emusic_silencing_sibilance/

[D-MAN]

It is definitely a recording artifact; the same thing that makes cymbals realistically "shimmer" and "sparkle" which is good for cymbals and triangles is a bad thing for vocalists. It is just bad mic'ing of vocalists, I think, usually indicative of being mic'ed too close.

On a "good" recording, it just doesn't happen for vocalists. Female vocalists are notorious for it, although Micheal McDonald of Doobie Bros. fame can even do it on his "T's" as well as his "S's" - not even Diana Krall can do that, as far as I can tell!

DM

[fini]

I saw Diana Krall live earlier this year. No problem with her T's.

[BEC]

bump

[LarryC]

Hmmm...."Sibilance" in itself doesn't seem to be a bad thing, as per the Am Heritage Dictionary definition below -- note that it refers to speech and related sounds rather than the high tingles of triangles and cymbals:

adj. -- Of, characterized by, or producing a hissing sound like that of (s) or (sh): the sibilant consonants; a sibilant bird call.

n. -- A sibilant speech sound, such as English (s), (sh), (z), or (zh).

So, aren't we concerned with whether the sibilance is accurately recorded, and then whether what is recorded (whether that's accurate or not) is accurately reproduced?

To me, an accurate recording or reproduction of sibilance is one in which the "s" sound is well detailed, without either (1) smearing and addition of distortion products or (2) exaggeration of frequencies in the sibilance range.

Note that sibilance, like cymbals and triangles, is a MASS of random frequencies packed closely together! This is the reason these instruments don't have a definite pitch -- they are a BIG BATCH of pitches. This is in contrast to pitched instruments like winds, strings, etc., which have comparatively simple pitch complexes.

It seems to me that a given level of distortion is greatly intensified and multiplied by such closely massed frequencies, and that a basically low, comfortable level of distortion that isn't obvious when playing pitched instruments can be made obvious by massed-pitch sound like vocal and instrumental sibilants.

So, it seems that if a system cleanly reproduces sibilance and cymbals that are well recorded, it's in good shape. At least I don't see how it can be blamed if it's called upon to reproduce poorly-recorded sibilance! That's why I think most will listen to the best available recording when trying to assess system reproduction, because that's probably the system's baseline quality.

Frequency exaggeration is another story. That can be speaker, crossover, amplification, CDP, cartridge, tonearm, turntable, etc., etc. Sibilance that is otherwise not that badly distorted can still sound bad, or tiring, if its frequencies, perhaps 1,000 - 3,000 Hz, are elevated. That's at least half the battle.

Larry

[Deang]

Good post.

It's certainly normal to hear it on some recordings, but I would say if you are hearing it with just about everything you play there is a problem. Listening to music should be enjoyable, not annoying or painful.

[Bennyboy]

Exactly ,

Some has told me you can't measure with your ears...................[]

Bennyboy

[DRBILL]

I, too, have a recording that I don't like but use for test purposes: Chet Baker Quartet "Love Nest" (cut 1) Pacific Jazz CDP 7243 8 55453 2 0. This features a trumpet solo with the trumpet fitted with a really acrid mute. Close microphony. It goes through you like a hot hairpin.

It is almost musical with the Crites tweeter but sounds like fingernails on a blackboard with the K-77.

DRBILL

[pauln]

Sibilance is present in the real sound of singing, which should sound natural, but really sounds terrible when there is too much. For those playing records, some sibilance excess can come from a stylus that has a resonant frequency in the 15-18KHz range which can create a band of high frequency hash that masks anything higher. This 'treble ceiling' in conjunction with a bass boost around 150Hz is the classic 'HiFi' sound of cheap consumer audio systems - boom and sparkle kind of sound. Similar sound comes from running full bass and treble on the tone controls. Tweeter response peaks might do this, too. Maybe some of the early CDs did this from some artifact of the very extreme lowpass filter employed to hide everything above 22KHz

A good stylus will be flat well beyond the audible range and should present the high voice sibilance sounds is a natural and realistic way - if the engineer was comeptent.

.

[oldenough]

One of my favorite artist's, Louis Armstrong, would seem to use this to good effect. Have a listen to Wonderful World, i hope this is not an unwanted artifact or bad recording, i have always enjoyed it partly because of the ssssibilance..

[ZAKO]

Sibilence begins with the mic. Its high energy almost distorts the first ununciation of a word or phrase. Alot of philes blame the speaker but its only the messenger. the proxemity to the mic can be changed or changing the mic itself will help. Some lazy recording eng. will just leave it in hopeing to change it in the mix. But by that time its too late.