PDA

View Full Version : "S for Sibilance" - a case study in deduction



pongal2000
02-02-2009, 05:05 PM
Hi,




This is my first post and a long one at that, and so please bear with me. I assembled my music system in bits and pieces from my travels. I live in Bombay.

My system consists of the following: Ayre CD player - LFD interconnect - LFD integrated - LAT Cable - Harbeth Compact 7 (ES3). I love my music system but for its tendency to sound sibilant. I also have a Project record player with a Goldring cartridge.

The CDs where sibilance is really high is a lot of the British Pop (Blur - Parklife/Cure - The Best of/Elbow - Leaders).
 Almost painful.
These CDs were new acquisitions.
 After noticing this, I started noticing this sibilance in several of my older music (Sarah Vaughn- Sarah Vaughn, Steely Dan ? Katy Lied (CD and LP) and Hoagy sings Carmichael), but not as much. Finally, I tried a speech LP (under the advice of Alan Shaw) ? Gerard Hoffnung at the Oxford Union, and the sibilance was low.

Though I liked the sound overall, I used to wince at the sibilance, and hence tried various other components. I tried another CD player (Quad) and that sounder slightly lower in sibilance, but not significantly so. I tried another amp (Nait 2, my older amp) that had somewhat lower sibilance too, but accompanied by loss of high frequency detail. In both cases, the trade off was that the audio excitement was lower.
I also tried a Shunyata power conditioner ? that made the sound much nicer, but the sibilance was still there.



I thought the Harbeth's tweeters may be a problem - 
I listened to the Harbeth LFD combo earlier, and it was lovely. So again under the advice of Mr. Shaw, I pulled out the grille and visually inspected the tweeter. No obvious dents. However, on several mono recordings, including the Gerard Hoffnung recording, the left speaker was noticeably more congested than the right, and had a tendency to rattle with some words. Both speakers were sibilant, but if I can put it as that, the left speaker?s sibilance was worse. This fact corroborated by my wife (who?s key message always had been: don?t be so hyper, it sounds fine). I interchanged the channels, and this problem continued with the left speaker.

The key difference between those shops and my living room is the absence of carpets. I live in a hot and dusty city and my daughter is asthmatic - so carpets are a no no. Furthermore, my speakers are in asymmetric settings: My living room is 11 ft by 18 ft (or so), but the right one (as I face it) has a dining room on its side (nearly the same size) and the left one has a small passage next to it, but to the side/ front of it is a wood wall (essentially the back of a cupboard), but for a brick column near the speaker. On the wood wall which is on the side of the left speaker, there is a large wall hanging backed by polyurethane foam. In front of the speakers (or behind the listening area) is a very large always open window.

Has anyone in the forum faced such problems? Type the words sibilance along with Ayre, Harbeth or LFD, and you?d hardly get a match. Any ideas on what I can do to fix the problem?

Thanks for your help,

Vivek

[Bold/underline emphasis added by AAS]

A.S.
02-02-2009, 07:05 PM
We've been corresponding off-group and I strongly suggested that this subject was brought here to the Harbeth User Group, so I'm glad to see it's here now. Together we may be able to work this through to a satisfactory conclusion.

Clearly if you "wince at the sibilance" something is very seriously wrong somewhere in your system. I have never, ever heard of such a gross defect with a Harbeth speaker and being easy-on-the-ears (as I've said many time before) is my No.1 design parameter. I wouldn't tolerate a speaker generating sibilance for more than a second or two so we do have to get you sorted asap! You've assumed that it's a problem with the speakers (and have offered to buy new tweeters) but I really don't want you to throw money at this without much more scrutiny, if at all.

There are lots of unknowns here, but as I understand it, what first brought this problem to your attention was when playing modern pop music. You didn't have this problem before then. Some older recordings (both on vinyl and CD) such as the Sarah Vaughan (1950s) and Hoffnung seem better to you. This is a little surprising because if the speakers were so defective that they were electrically or mechanically damaged (for whatever reason, accidental or otherwise) then I think that all recordings would sound sibilant especially on vinyl, where surface clicks and crackles have a very wide frequency content, far more so than CD which cuts-off sharply at 22kHz. Also you say one speaker seems more problematic that the other.

So, we have a few facts which lie uncomfortably together, including your note that your listening room is asymmetric and that you have a relatively hard, untreated environment for good reasons. I really don't think that a room could add what you describe as sibilance to an otherwise perfectly normal hi-fi system; the room could add a brightness or even a hardness but that is definitely not what we'd call sibilance.

Analysis Step 1:

So the first step is to be really sure that you are using the word sibilance correctly. Sibilance is described as "Of, characterized by, or producing a hissing sound like that of (s) or (sh)". Please make those 's' or 'sh' sounds out loud to identify them and confirm to me that those sounds are being emphasised in your system and not any other vocal sounds - only 's' and 'sh' or 'chrrrr'. This is important. These are sounds that can only be generated by blowing air at pressure through the teeth. They are absolutely distinctive.

Gan CK
03-02-2009, 01:34 AM
Hi,




This is my first post and a long one at that, and so please bear with me. I assembled my music system in bits and pieces from my travels. I live in Bombay.

My system consists of the following: Ayre CD player - LFD interconnect - LFD integrated - LAT Cable - Harbeth Compact 7 (ES3). I love my music system but for its tendency to sound sibilant. I also have a Project record player with a Goldring cartridge.

The CDs where sibilance is really high is a lot of the British Pop (Blur - Parklife/Cure - The Best of/Elbow - Leaders).
 Almost painful.
These CDs were new acquisitions.
 After noticing this, I started noticing this sibilance in several of my older music (Sarah Vaughn- Sarah Vaughn, Steely Dan ? Katy Lied (CD and LP) and Hoagy sings Carmichael), but not as much. Finally, I tried a speech LP (under the advice of Alan Shaw) ? Gerard Hoffnung at the Oxford Union, and the sibilance was low.

Though I liked the sound overall, I used to wince at the sibilance, and hence tried various other components. I tried another CD player (Quad) and that sounder slightly lower in sibilance, but not significantly so. I tried another amp (Nait 2, my older amp) that had somewhat lower sibilance too, but accompanied by loss of high frequency detail. In both cases, the trade off was that the audio excitement was lower.
I also tried a Shunyata power conditioner ? that made the sound much nicer, but the sibilance was still there.



I thought the Harbeth's tweeters may be a problem - 
I listened to the Harbeth LFD combo earlier, and it was lovely. So again under the advice of Mr. Shaw, I pulled out the grille and visually inspected the tweeter. No obvious dents. However, on several mono recordings, including the Gerard Hoffnung recording, the left speaker was noticeably more congested than the right, and had a tendency to rattle with some words. Both speakers were sibilant, but if I can put it as that, the left speaker?s sibilance was worse. This fact corroborated by my wife (who?s key message always had been: don?t be so hyper, it sounds fine). I interchanged the channels, and this problem continued with the left speaker.

The key difference between those shops and my living room is the absence of carpets. I live in a hot and dusty city and my daughter is asthmatic - so carpets are a no no. Furthermore, my speakers are in asymmetric settings: My living room is 11 ft by 18 ft (or so), but the right one (as I face it) has a dining room on its side (nearly the same size) and the left one has a small passage next to it, but to the side/ front of it is a wood wall (essentially the back of a cupboard), but for a brick column near the speaker. On the wood wall which is on the side of the left speaker, there is a large wall hanging backed by polyurethane foam. In front of the speakers (or behind the listening area) is a very large always open window.

Has anyone in the forum faced such problems? Type the words sibilance along with Ayre, Harbeth or LFD, and you?d hardly get a match. Any ideas on what I can do to fix the problem?

Thanks for your help,

Vivek

[Bold/underline emphasis added by AAS]

Hi, welcome to HUG. Its hard to imagine that the LFD + Harbeth C7 ES3 combo can give rise to sibilance. If you ask me, i'd think the pop cds you just acquired might be the main culprit. In addition, the LFD is a highly transparent & revealing amp & as such will not mask any shortcomings in modern recordings. I too face lots of sibilance when i play some china pressed cds. There's plenty of 'shh shh' in the vocals but i am not perturbed by it at all because i know its in the recording. The other probability might be your cables. Try to loan some other cables to see if the problem persists. Btw, what stands are you using? If you are using metal stands, try to get it filled or change to wooden type if possible. Hope this helps.

P.C.
03-02-2009, 02:44 AM
Hi Vivek, just an idea but maybe you could try and borrow a headphone amp and good headphones and listen to the bad cd's . At least this will remove any room bound problems that you could have. If the music sounds fine it could still be the amp/ speakers or room. but if its still sounds sibilance you'll at least know its not the speakers or room that's causing the problem.

skylan
03-02-2009, 06:36 AM
Hi Vivek,

If you hear more Sibilance in the left speaker then move it over to the right and vise versa. If there is still more Sibilance in the left speaker at his point, then the issue may be more dedicated to the components or interconnects.

On another note.

The compact-7 likes a stand of 20" high and also if you fold a blanket in half and place it on the floor in front of each of the stands, it will help to reduce first reflections,(Bounce) from the smooth hardwood.
A carpet acts as floor treatment.

Noel.

pongal2000
03-02-2009, 08:35 AM
Hi,

Thanks for your inputs.

To answer the questions (all except Mr. Shaw's):

a) I use wooden stands which I had made in India. They are made of MDF (on top) and speaker placed on using a small amount of Blue Tac.

b) On the Brit Pop recordings, using my Apple Laptop and Sennheiser PX100 headphones, the sound is a lot less sibilant. I will listen again to reconfirm.

c) On the individual speakers per se, my wife and I listened at some 0.5m from the speakers - almost like using them as a headphone! In essence, I was trying to eliminate as much of the room related issues as possible.

d) I will also try the blanket suggestion.

I will answer Mr. Shaw's questions after another listening session by this weekend!

Regards,

Vivek

T.W.
03-02-2009, 01:14 PM
Just some more thoughts:

It may be a temperature issue.

It may be dust issue.

The electricity supply may be extremly bad where you live. If you listen with your laptop
then you probably use the battery, so here it's not an issue. If I don't use a power conditioner,
then there is sometimes some light sizzle. Conditioners might help. If electricity is an issue
then try to borrow a PS Audio Power Plant Premier - it recreates everything and is extremly
good but expensive.

Try another electrical outlet. Use just ONE for the entire equipement.

Switch the speaker by just changing the cables. Use other, simple cables to test.

Move the speakers very close together and use the ballance control (if you have one)
in mono (if you have it) and compare both speakers.

Look into your CD player if the laser is dirty. Clean it with alcohol. (I openened my Marantz
CD player once a year ago - oh boy - unbelievable!)

Move your stuff to a friend and try it there.

Good luck,
TW

A.S.
03-02-2009, 02:11 PM
Yes, it could be all or any of these things, but frankly, for this user to say that he "winces at the sibilance" (his words) there must be a chronically serious problem somewhere in the system, far beyond that of bad cables, power conditioning, stands or dirt and dust. In my book, sibilance is as bad as it gets: there really isn't anything more fatiguing or less musically satisfying than sibilance.

After suitable deduction, we may conclude that something is indeed wrong with the speakers. We here can resolve that. My concern is that Harbeth speaker failures are so extremely rare that statistically the odds are that the problem is somewhere else in the system. To my knowledge, I can't recall a single C7ES3 tweeter failure; they're well made and designed and we test every one of them.

May I suggest that until we have some detailed, constructive feedback (answering my question posed in Point 1) that we just hang fire - please. We don't want to put further concerns in this users mind. As I read it, this sibilance problem didn't exist until he tried playing modern pop music on his speakers.

P.S. One thing to add. The User mentioned that we have a clean, sibilant-free speech test recording downloadable here recorded (by Derek and myself) inside the BBC's anechoic chamber with a perfect microphone. There is no hint of sibilance on this recording. Indeed, deprived of all the brightening effect of sonic echoes off normal hard surfaces, it may surprise you how 'dull' our voices sound. But this recording is technically near-perfect and should prove a point: if you hear sibilance on our recorded voices on this clip there is something wrong with your system somewhere. It maybe the speakers. It maybe the electronics. It maybe your hearing (unlikely) but it's definitely not this recording.

Harbeth Reference Speech Recording download. (http://www.harbeth.co.uk/usergroup/showthread.php?p=2265#post2265)

>

tinears02
04-02-2009, 01:09 AM
Vivek,
Can you borrow another pair of speakers to compare? Even a non-Harbeth that you
are familliar with should do, since the sibilance you mentioned is so apparent.
Some bad connection or dry solder joint in the high current signal paths
may be a culprit too.

yh

pongal2000
07-02-2009, 12:45 PM
Hello,

This is in reference to the test for sounds.

On first hearing of a few CDs, the sounds that are most emphasised are ?s? (including the soft ?c? in some songs), and to some extent ?sh?. Sounds like ?th? (used in ?through?) or ?t? (?turtledoves?) come across as a bit spitty. The letter ?f? ("further") also tended to sound a bit prominent and spitty.

Unfortunately, I could not download the recording mentioned by Alan Shaw. It only keeps playing in my laptop using Quicktime - if any Apple user has suggestions on how to download, please do let me know.

Lastly, yh, thanks for your suggestion - will try and get another speaker to test, but that is going to be a difficult one (not too many shops nearby and have been working long hours)!

Thanks and regards,

Vivek

A.S.
07-02-2009, 06:38 PM
OK that's useful feedback.

Now I don't know anything about MACs (I'll call my youngest son, David, who is a big Apple fan and report) but as I advised, did you right click and 'save target as' - or is that only something applicable to PCs?

It's critical that you play this speech test. We can't begin serous further analysis of this case until you have done so.

DAVIDSHAW
09-02-2009, 09:37 PM
I have successfully downloaded the MP3 file on my Mac. I did this by:

1) Pressing 'ctrl' and pressing the mouse pad then choosing 'Download Linked File As' and choosing the location to save the MP3 to (typically the desktop).

2) Alternatively, if you are using a plug-in mouse or your Mac laptop has the multi-touch mouse pad enabled, you can right click and choose 'Download Linked File As' - as above.

The MP3 file can then be played in iTunes or QuickTime (or any other software for that matter).

Hope that sheds some light on this issue.

Regards,

David

A.S.
09-02-2009, 09:50 PM
Thanks son, but how do you get it from your Apple computer into a WAV format and burned to CD to play on your hi-fi CD deck?

DAVIDSHAW
09-02-2009, 10:06 PM
In order to convert the MP3 into a WAV file (after downloading as instructed above), please follow these instructions (on a Mac):

1) Open iTunes > click on 'iTunes' (top left) > 'Preferences' > Stay in the 'General' tab > 'Import Settings' > 'Import Using' and choose WAV

2) Staying in iTunes Preferences' > click on the 'Advanced' tab (far right) > and tick the box 'Copy files to iTunes Music folder when adding to library' (this will ensure the MP3 file is saved into the iTunes music library in the next stage).

3) Drag the downloaded MP3 file from its saved location (preferably the desktop for ease) into the iTunes Music library (the music note symbol in the left side column). A green + symbol will appear (indicating the file is being imported).

3) Once it has copied itself into iTunes > use the search tool (top right) > type in 'Alan' and it will find the MP3 file.

4) Select the MP3 file and right click or press 'ctrl' and press the mouse pad, and choose 'Create WAV version'

5) Once this is done, there will be two files with exactly the same name. One will be the original MP3 and the other will be the newly created WAV. As above, select the WAV file and right click or press 'ctrl' and press the mouse pad, and 'Add to Playlist' of your choice.

6) Once the WAV is in a playlist (created on the left column), you can right click or press 'ctrl' and press the mouse pad on the playlist and choose 'Burn Playlist to Disc'. You will be prompted to insert a blank CD and iTunes should do the rest.

You're then ready to use this CD with your Hi-Fi.

Regards,

David

pongal2000
15-02-2009, 04:28 PM
Hello again,

First of all, many thanks to David for facilitating the download.

Going on, there are a couple of things that I have done, which I thought I would highlight before I get to the speech recording.

To start with, I took one of the sibilant CDs (Hoagy Carmichael) to a friend?s ? he has an entirely different hi fi comprising of pretty well known names. The CD sounded sibilant in his hifi as well, but possibly less. However, there seemed to be some trade off ? there was a lot less air around the vocals (gosh, I think I have been looking at too many audio reviews!). The presentation through his music system was very different from mine though. I don?t know if that makes any sense, but it sounded more like listening to a lovely hifi rather than listening to music live.

Secondly, I changed my cable around. I got a LFD cable based on the recommendation of a really nice audio dealer out of the US who carries the same lines (someone whom I trust since I the time I bought my first system 20 years ago). The sibilance was much reduced ? it was still there, mind you, but a lot less unpleasant. The British pop CDs (Cure, Elbow (I wonder if Rebus and Siobhan also noticed the sibilance on the latter?! They are the only two other whom I know listened to the album - anyone else in this group? Comments? ) were still sharply sibilant, but it was such that it did not rob the impact of the next word (earlier, it would be a series of annoying sibilant sounds ? ugh!). Sarah Vaughn ? much, much less in sibilance. Acceptable! I also heard Requiem (Mozart ? Gardiner) ? fabulous, though some natural sibilance in Dies Irae.

Finally, the speech test CD. Alan?s voice came across as sibilant (for example, around the Twenty and Seven in the first line itself). Derek?s voice was completely devoid of any sibilance. The same was observed using my headphones as well. My wife listened to it as well (I am amortising goodwill) and I have asked her to type in her comment which is as follows ?the sibilance is there, but it seems natural. Some people are more simply more sibilant than the others?.

Hmm?all this should lead to a logical concluding sentence, but I cant think of it. So I have just left the facts be and open for interpretation. Alan, you have blamed modern recordings elsewhere. Do you have some music recordings that you use as a reference for natural sound?

Vivek

A.S.
15-02-2009, 11:31 PM
...Finally, the speech test CD. Alan?s voice came across as sibilant (for example, around the Twenty and Seven in the first line itself). Derek?s voice was completely devoid of any sibilance. ... ?the sibilance is there, but it seems natural. Some people are more simply more sibilant than the others?.Your wife is absolutely correct. I've just listened to the recording again and fundamentally, there is more 'top' in my voice than Derek's (I suspect his beard soaks up high frequencies!) but really, there is no unnatural sibilance in either of our voices, just a normal difference in balance.

Sibilance is really an over-exaggeration of high-frequency energy, and you can't convince yourself that any human anywhere could produce that much 'top' in their real, live, unrecorded natural voice. Sibilance is akin to the the ringing of a high-pitched bell or a wet finger on the rim of a wine glass superimposed onto the natural sound. That's why sibilance is fatiguing if your hi-fi system exhibits it - your subconscious says 'a human can't make that horrible noise so by definition I'm not listening to a real human ... I must be listening to a reproduced voice'. Generally speaking, many or even most modern recording are too toppy, and most microphones have a high-frequency boost and are placed too close to the performers which emphasises the higher tones. It should be no surprise that most modern CDs are indeed too bright and what you might (incorrectly?) attribute to being sibilant.

So, assuming that the system passes the Harbeth speech test using our anechoic recording with ruler-flat microphone, and allowing for the fact that there is a spectral difference between Derek's and my voice, and there is no dominant ringing, then I don't think that any of my pet music recordings are going to prove anything. As Pluto says here (http://www.harbeth.co.uk/usergroup/showthread.php?p=4495#post4495) speech really tells you a huge amount about the performance of the speakers because we all innately know what a human voice box sounds like even if we have never heard a real, live musical instrument. I'm not quite sure what to suggest next other than swapping the speakers for something else as suggested in an earlier post. Would you think I was extremely rude in suggesting that as a scientific exploration you consider having your hearing tested just in case you have always had or have lately developed an enhanced sensitivity to high frequencies? It can and does happen and we should eliminate this avenue of investigation. It would certainly put your (and my) mind at rest and then we can start to consider everything else.

I'll pass on your best wishes to David who is on his university third year, currently on industry-placement year with Fujitsu and very much enjoying it. My first real job was with NEC Corp. so I'm pleased that history has repeated itself and that we've both had the chance to work for mighty Japanese corporations. There is no better training than working for a Japanese company who bring products to market with textbook precision: I will be forever grateful to NEC.

P.S. If you listen to AM radio - that is, medium wave or short wave radio, how does it sound to you?

pongal2000
23-02-2009, 02:43 PM
Hi Alan,

No it would not be rude to suggest to get my ears tested (though my wife would insist that I do the rest of the head along with it), it would be fun. But I will leave it for another day.

That said, just to make sure that what I am listening to is not just me, I double-checked everything with my wife (who has been my listening partner through the years of "auditions"). I made sure, as much as possible, not to condition her into what to expect.

To answer one of your other questions, I dont have an AM radio.

I think I am the end of this journey. I listened to your speech test with Spendor 3/5s and the sibilance was not as pronounced, but still quite there. They were no match for the Harbeths in terms of dynamics and extension. On the whole, I am now happy to live with the Harbeths. I also discovered two recordings along the way where sibilance is low ? ?Walking on the moon? by The Police and ?When I am sixty four? by the Beatles.

I must say that swapping cables have made the single most dramatic difference ? to recap, I switched to the LFD cable a few weeks ago.

Finally, being a long weekend here, I went and got some polyurethane and polyethelene foam pieces and strewed it along sensible points (to the wall next to the speaker, on the floor at the first reflection point and across to the other side of the wall). I also used wood fibre panels behind the speakers which I thought would act as a diffuser. The effect on the sound was magical. Imaging was amazing, vocals popped out more prominently, etc. The room (clap test) was also less reverberant. The sibilance was there, but not bothersome.

Vivek

pongal2000
01-07-2009, 06:31 AM
Hi,

Sorry to revive an old post, but there is a happy ending.

I changed a component along the chain in my system (not so much cable and stands). The whole sound changed into something brilliant and incredibly satisfying - transparent, but without fatigue. The sibilance is only the natural one that is there on certain recordings. Even in the "Best of Cure" CD that I alluded to, the sibilance is still there, but not at the cost of the music (many details have emerged out of the background).

I am having a wonderful time listening to music - gosh, I didnt know that disc 1 of the "White" Album (The Beatles) was so brilliant! The bass solos in "Way Out West" come out very nicely...(well, by now the drift is clear).

Separately, there was another post that looked at best discs for listening to music systems. Can I recommend that (a) please take poorly recorded, but music that you listen to everyday, rather than well recorded hi fi records alone. That will be a good test of how a bulk of the music sounds in the system and (b) download the voice conversation that features Alan - very useful.

Regards,

Vivek

egoBen
26-01-2013, 05:24 PM
Hi,

Sorry to revive an old post, but there is a happy ending.

I changed a component along the chain in my system (not so much cable and stands). The whole sound changed into something brilliant and incredibly satisfying - transparent, but without fatigue. The sibilance is only the natural one that is there on certain recordings. Even in the "Best of Cure" CD that I alluded to, the sibilance is still there, but not at the cost of the music (many details have emerged out of the background).

I am having a wonderful time listening to music - gosh, I didnt know that disc 1 of the "White" Album (The Beatles) was so brilliant! The bass solos in "Way Out West" come out very nicely...(well, by now the drift is clear).

Separately, there was another post that looked at best discs for listening to music systems. Can I recommend that (a) please take poorly recorded, but music that you listen to everyday, rather than well recorded hi fi records alone. That will be a good test of how a bulk of the music sounds in the system and (b) download the voice conversation that features Alan - very useful.

Regards,

Vivek

I know it's a very old thread. But if Vivek's around, I'd like to know what was changed and for what. The speakers itself?

Miles MG
20-07-2013, 03:44 PM
Sibilance. It's a common problem with inexpensive electret mics. I do a certain amount of mainly ( speech ) recording and females with naturally sibilant voices are slightly emphasised by the Sony electret mic. I use in conjuction with a small portable recorder.

Deeper female voices are OK with this microphone, so it's a case of ' horses for courses.' For serious recordings I employ mainly a ribbon or an expensive condenser, such as a Neumann. Monitoring using my small Harbeths, these certainly expose the quality of the microphones used.