Jump to content
Maestronet Forums

Can Youtube movies learn you something about violin sound?


Anders Buen
 Share

Recommended Posts

Are there any other programs which are less expensive or is this the best one for the job?

I have an Apple and SpectraPlus is Windows based, my husband has a Windows laptop but I would like a program that could be put onto either computer.

He is interested also as he is looking to making reeds for a concertina shortly so we could share a program hopefully.

I would love a program that is easy to use :)

Some are just so technical it is hard to find where to start. :)

Thanks.

Sharron

Sorry for the late response but i hit UPLOAD to my first response, so hopefully i do it right this time. I bought a software program about 8 years ago called Sound Snip that i just recently tried and i think it might help you for it is pretty easy to use. I bought it from E_glass for around 60$ and a mike from radio shack for around 50$ to input the sound into the computer program. I use it to get a paper record of the frequency compostion of a top or back plate while i'm thinning it to see what chnges occur. If i tap on an assigned spots, holding the plate by the lower bass corner the program will display the cumulated frequency of tonal notes/hz up to 5000hz. Simply tapping a pencil on various spots i've selected while i'm thinning somewheres i can get a fairly good duplication up to 1500hz, and improving the tap method should allow me to get to 2000hz which covers a large part of the sound of an inst. What i do is print out the first tap and then repeat the tap comparing the screen displayed tap to the print out, and if the match is ok i'll save it. I can show a sample of the print out if someone is interested and also how to go from scanner to Maestronet. Obviously such data only has meaning to the maker, for if a paper record of the taps of an assembled inst turns out that the inst is a stinker with a magnificent wolf note, you'll certainly try for a different combination of taps. I've only tried this on a few spare tops i made and one i removed from an inst and each one is like a fingerprint, very diffeerent. I just wish i tried it when i first purchased it. fred

Link to comment
Share on other sites

  • Replies 108
  • Created
  • Last Reply

Top Posters In This Topic

Sorry for the late response but i hit UPLOAD to my first response, so hopefully i do it right this time. I bought a software program about 8 years ago called Sound Snip that i just recently tried and i think it might help you for it is pretty easy to use. I bought it from E_glass for around 60$ and a mike from radio shack for around 50$ to input the sound into the computer program. I use it to get a paper record of the frequency compostion of a top or back plate while i'm thinning it to see what chnges occur. If i tap on an assigned spots, holding the plate by the lower bass corner the program will display the cumulated frequency of tonal notes/hz up to 5000hz. Simply tapping a pencil on various spots i've selected while i'm thinning somewheres i can get a fairly good duplication up to 1500hz, and improving the tap method should allow me to get to 2000hz which covers a large part of the sound of an inst. What i do is print out the first tap and then repeat the tap comparing the screen displayed tap to the print out, and if the match is ok i'll save it. I can show a sample of the print out if someone is interested and also how to go from scanner to Maestronet. Obviously such data only has meaning to the maker, for if a paper record of the taps of an assembled inst turns out that the inst is a stinker with a magnificent wolf note, you'll certainly try for a different combination of taps. I've only tried this on a few spare tops i made and one i removed from an inst and each one is like a fingerprint, very diffeerent. I just wish i tried it when i first purchased it. fred

Fred,

Could you find a link for the software as I can't find it with a search.

If it has been around for 8 years maybe the name has changed or is no longer available.

I am not worried about having the latest gadget, I just want something simple to use that does what it says on the box.

Sharron

Link to comment
Share on other sites

The analysis was done using a 1/3rd octave band analyzer connected to a CD portable player. The left and right signals are by the way slightly differnt also in the frequency components.

The high frequency noise on that recording is also my fault I think. I ran the recording through a casette recorder in other to be able to hear the signal at the same time.

this makes me laugh. you try to analyse frequency response on unknown recordings but run them through your CASSETTE recorder first?

unbelievable. from high tech to stone age in just a second. i wish you fun with your results.

Link to comment
Share on other sites

this makes me laugh. you try to analyse frequency response on unknown recordings but run them through your CASSETTE recorder first?

unbelievable. from high tech to stone age in just a second. i wish you fun with your results.

If you take a look and compare the two figures in the first post you will se that the one gone trough the casette recorder amplifier and the one wich has not are quite similar for the high frequncies. Compare the black curves. The explanation for the high 200Hz band level in the first figure is explained in another post under this thread. - You must be sleeping! - Or just not getting the points!

Anders

Link to comment
Share on other sites

Anders,

Thanks for the LTAS analysis, it is interesting to ponder.

Is there a way to do the 1/3 bands with math?

This is different than the full fft analysis that shows more a signature of each violin's voice. I am failing to see much above 400 Hz as all the lines seem to cross.

I have not tried to do the 1/3rd octave band analysis with math. However it is done in the programs that do the job. Once these filters were analog circuit filters with rounded corners and hills at each side with a final steepness. I guess it could be done by calculating the pressure from the dB scale fft, sum the pressures in the band regions and then recalculate that sum by taking the 10*logarithm to that sum. I guess the fft spectrum show the sund pressure levels squared. It might be easier if you start off with a linear fft spectrum and not a logarithmic one.

I think the bandwidth is 0.25 times the 1/3rd octave centre frequency.

Link to comment
Share on other sites

If you take a look and compare the two figures in the first post you will se that the one gone trough the casette recorder amplifier and the one wich has not are quite similar for the high frequncies. Compare the black curves. The explanation for the high 200Hz band level in the first figure is explained in another post under this thread. - You must be sleeping! - Or just not getting the points!

Anders

no, i don´t get the points of analyzing recordings from unknown sources which have been run through unknown conversions of unknown resolutions. don´t get me wrong - what i´m trying to say is: get your source material sorted out before you go into analyzing it. think about AD and DA conversions for instance. how many times have you converted your source, and how many times has it been converted before you got hands on it? look, i am a sound engineer by profession, and one of the "golden" rules is "once you´re digital, stay digital". did you work according to this? bet you didn´t.

Link to comment
Share on other sites

Fred,

Could you find a link for the software as I can't find it with a search.

If it has been around for 8 years maybe the name has changed or is no longer available.

I am not worried about having the latest gadget, I just want something simple to use that does what it says on the box.

Sharron

Hi Sharron

I checked the Southern Cal Asso of Violin Makers and he is still a member and his phone is 310-390-0761, email is e_glass@yahoo.com. His name is Edwin Glass. The microphone i use is Radio Shack 33-3013. It is less than 1/2 inch in diam and will fit into an ff hole of a viola. It has a small battery to amplify the sound. I hold the object i tap about 6 inches away and hope my refrigerator doesn't start up. I swear it runs all day. I'm kicking myself for not trying it when i first got it. If i had a paper record of of the dozen or inst's i've made it might have helped, or at least i might have a little more understanding of getting certain frequencies in various parts of an inst. Please contact me if i can be of any help. fred.

Link to comment
Share on other sites

no, i don´t get the points of analyzing recordings from unknown sources which have been run through unknown conversions of unknown resolutions. don´t get me wrong - what i´m trying to say is: get your source material sorted out before you go into analyzing it. think about AD and DA conversions for instance. how many times have you converted your source, and how many times has it been converted before you got hands on it? look, i am a sound engineer by profession, and one of the "golden" rules is "once you´re digital, stay digital". did you work according to this? bet you didn´t.

This must be the wrong forum for you! I am an acoustician by profession. I do not learn anything from what you are saying.

Link to comment
Share on other sites

This must be the wrong forum for you! I am an acoustician by profession. I do not learn anything from what you are saying.

..................................................................

Anders, please could you be very exact about your acoustic professional credentials...I think you should if you wish to take a stand like this.....i.e. please state your academic qualifications, professional specializations and your professional status in your company.

Thanks.

Link to comment
Share on other sites

This must be the wrong forum for you! I am an acoustician by profession. I do not learn anything from what you are saying.
..................................................................

Anders, please could you be very exact about your acoustic professional credentials...I think you should if you wish to take a stand like this.....i.e. please state your academic qualifications, professional specializations and your professional status in your company.

Thanks.

I'm more curious about what you plan to do with data obtained in this manner? If it's just for putting onto websites like this then it's not too bad. I just can't imagine trying to do any kind of serious work with recordings made by equipment that you don't know very well.

For the past year I have been building an instrument to measure the acoustic impedance of different violin f-hole shapes. I've built four versions of my testing apparatus to get to the point that I finally feel like I can take a good measurement. I just have to improve a few gaskets and slightly modify my f-hole mounting scheme and I am finally ready to take data, after over a year of work building and understanding my equipment. Of course this could be done quicker if I was actually doing acoustics for something other than just fun in my spare time.

Link to comment
Share on other sites

..................................................................

Anders, please could you be very exact about your acoustic professional credentials...I think you should if you wish to take a stand like this.....i.e. please state your academic qualifications, professional specializations and your professional status in your company.

Thanks.

Who is writing? - I expect that you give your own credentials.

I am a civil engineer in experimental physics from 1993 wich is equivalent to a master of science (at least the university I attended regards it tat way). I took my project work and engineering thesis on vibration modes in violins using TV-holography and started reading publications on musical acoustics. I was reading literature on the violin in particular during that years work and have continued ever since. I have a good collection and insight into the literature on the subject. I have some courses in acoustics (physics deals with acoustical phenomena on a fundamental basis, quantum mechanics also use phenomenology from the acoustics, so there are some analogies. I also have some course in optics wich also has analogies to acoutical phenomena) from that study and have taken several since in my now 8+ years as an acoustics consultant with Brekke & Strand akustikk as in Oslo Norway. We are 25 acousichians working together in (one of) the leading acoustics consultant companies in Norway and I am e.g. a member of the acoustics consultant team for the new national Opera House in Oslo wich has now just been opened.

I have a course in musical acoustics from Udine in 2006 (CISM), three years at the VSA oberlin violin acoustics workshop (the first year as an invited speaker). I have attended more than 15 acoustics conferences and read the JASA and Acta Acoustica regularly.

I come from a family of folk musichians and makers, the musicians dating back several hundred years. But makers back to about 1915. They were farmers but ran a violin and hardanger business like I do now bseides that. My uncle is both a player and maker and some of the best players today own an instrument made by him. I built my first hardanger of age 13, some 25 years ago. I consider my 'acoustics career' to start from there. My father runs a record company for folk music, so I grew up with recording equipment and sound egineers around. Some of my best friends are sound engineers and are good at it.

We do a fair amount of measurements. My expertise is in the combination of modeling and measurements, both in room- arcitectural- and some environmental acoustics. We do auraliations based on calcuations and measurements. I deliver 4 papers to the IOA Auditorium Acoustics conference now in Oslo in early october and is one of the main speakers there. (Two papers is with others, two are written by me)

I think I am one of the leading workers in our company, there is lots to do. Last year the founder and I had the best turnover. And I am probably the most active publisher besides a collegue of mine running the http://www.akutek.info/index site and my boss wich is a PhD with speciality on hearing. He has several JASA papers on investigations on hearing on a large population in the mid norway (50000 persons, the biggest stydy ever done on humans). I have no publications in a serious peer rewieved acoustics journal besides the CASJ and VSA Papers.

All my 'rough science' work is done without founding. Its done in my spare time.

I am a member of the Acoustical Socisty of America, VSA, The Norwegian Acoustical Society, the Norwegian Physics Society and the Norwegian association of civil engineers.

I am sort of an creative 'rough scientist' with somewhat broad skills in acoustics. I get my bread and butter from the consultant work.

Link to comment
Share on other sites

Interesting anders.

It seems that the road from engineering to acoustics is a common one.

Do you play or make violins ?

Nigel Harris was also a civil engineer who later became a self taught violin maker.

He has quite a P.H.D. in 'violin related' acoustics, & his fiddles sell for $16,000.

My Step father was the head of the civil engineering dept at the BBC for 20 years,

and yet he's probbaly the most impractical person I have met, & clumsy as hell.

His idea of life is a theory, and half of it doesn't even exist.

So, my point is that it takes all sorts to make the world go round.

Cheers.

Thanks!

Yes, I play the hardanger, and the violin a little. Folk style playing and much self taught. I do not perform.

I make the hardanger and regraduate some violins. Low prouction though. Some repair work.

I am interested in the process sound engineers are doing and know that e.g. the sound and acoustics on bradcastings and recordings my be manipulated a fair bit if not totally by sound engineers. Added reverberation, pull up or down certain voices or instruments, manipulating their timbre etc. If there is no sound engineer between the mic and the recording device you are a bit safer..

Anyway, still the process of analysing sounds from recordings are interesting in my opinion because the same material is available to many and you can compare what you hear to what you get out of the analysis.

Besides that there is a fundamantal discussion between sound engineers and acoustichians because the latter uses simplifications that seems odd to the recording engineers and vice versa. I think there is an interesting fundamental discussion that can be taken about statistical view on things (acoustichian, averaging room acoustics in octave bands) versus the fundamental point response high resolution thing that sound enigineers and academics seem to follow. There are adavenatages to both sides. I might generalize a bit too much here, but still I hope some see my point.

There are some similarities between how a sound engineer works and how we listen ro rooms using auralization e.g. Some of the most important things I have learnt this year was from a sound engineer that taught me about Altiverb and how he was working with sound production.

I have a lot to learn and do have respect for that profession. There is also a tendency to take details very serious wich proably does not matter much to the hearing. Therefore an acoustichian sometimes will be sceptical about some ideas and theories for sound production sound engineers may have. However there is some analogies between what they writen in HI-Fi and audio engineering magazines and what we think about violin sound.

For an aoustichian recording equipment and reproduction systems should have a flat response, reproducing the sound naturally. That is an attitude wich does not include the sound engineer in the measurement setup.. Very exclusive!

That became more text than I intended. I really should work now.. :-)

Anders

Link to comment
Share on other sites

I've always been fascinated by this type analyses and believe it could be a very useful tool. Michael worked with it a few years ago and gave me a few links for trial software; alas I never had the time to pursue it.

My experience and knowledge is likely not as good as many here but I do have one thought to add. Do you not think that the strings used would be a significant wild card in the you tube comparisons; particularly when analyzing the older masters. To my knowledge they played on unwound gut strings which should favor low frequency sound. I believe most of them did switch to a steel e string for dependability which could explain the high frequency similariities.

Just a thought and thanks for sharing your work.

Cheers!

Link to comment
Share on other sites

I've always been fascinated by this type analyses and believe it could be a very useful tool. Michael worked with it a few years ago and gave me a few links for trial software; alas I never had the time to pursue it.

My experience and knowledge is likely not as good as many here but I do have one thought to add. Do you not think that the strings used would be a significant wild card in the you tube comparisons; particularly when analyzing the older masters. To my knowledge they played on unwound gut strings which should favor low frequency sound. I believe most of them did switch to a steel e string for dependability which could explain the high frequency similariities.

Just a thought and thanks for sharing your work.

Cheers!

My experience is that steel strings give more low frequency response thnan does gut strings. Hardangers use gut G, D an occasionally A strings. Thin though, but violin strings e.g. Dominants tend to give more sound, also in the low frequencies than gut strings. But I really havent tested gut violin strings in a manner to draw conclusions. My impression is that the higher low frequency response of the older violins is a 'violin body and construction thing' and not something to do with the strings. Chinese violins I have bought come with steel strings that improve the low frequency response compared to e.g. Dominants..

Anders

Link to comment
Share on other sites

What happens if you use the same player, the same instrument, the same environment, and two differnt type of strings ?

I mean, can the computer thing tell the diference between two synthetic types like Dominant and Pirasto ?

Cheers.

I have tried this using the steel strings and Dominants on a chinese violin. I have not tried Dominant versus Pirastro so I do not know. I am really not very interested in strings. I have tried a set of Zyex on one I used to have Dominants on and they seem to ring a bit more in spite of being more damped according to the producer. I have response plots of these two situations but not from the same room I am afraid. Seem to be a bit more powerful, but not rally sure.

Anders

Link to comment
Share on other sites

If you are going to be doing any kind of meaningful analysis on recorded violin sound, a flat microphone must be used such as a AKG C460 or an equivalent flat response small diaphragm reference condenser mic. Then the instrument must be played in an anechoic chamber to eliminate room reflections and equalization, and then recorded using nothing more than a flat response preamplifier feeding a flat response recorder, preferrably digital based. This creates a repeatable benchmark if the instrument(s) are always played the same distance from the microphone by the same musician.

Violins like most things sound terrible in an anechoic chamber, but it reveals the true sound of the instrument uninfluenced by reverb, comb filtering, and other things that a reflective room will do to the sound.

A Radio Shack microphone, cassette recorder etc. may be used for some very rough comparisons, but I doubt the results will offer much in the way of useful information. As for evaluating violins on Youtube, you are at the mercy of the sound engineer that recorded it. A good engineer can make a lousy instrument sound somewhat acceptable, and a mediocre violin sound quite good through EQ, mic selection and placement, reverb, etc.

It is for this reason I don't put much stock in sound clips of instruments on the internet. They may be interesting to listen to, but I think there is little to be gained save for entertainment purposes.

Link to comment
Share on other sites

I am really not very interested in strings.

My point is that if the computer can recognize a difference between two brands of nylon strings, then perhaps it might be able to detect what a trained ear can hear.

Otherwise, surely you're just talking decibels, not tone.

A good set of ears can detect nuances that the best microphone in the world cannot accurately capture.

Link to comment
Share on other sites

What happens in the space between the ears is just as important as what happens in them.

True that rigorous methods for recording and testing instruments is best, but the major resonances are quite robust and identifiable.

Even though different types of strings are used the basic acoustical structure of the instrument remains the same. The point is what the soundbox does with the vibrations from the strings.

~OK

Link to comment
Share on other sites

What happens in the space between the ears is just as important as what happens in them.

True that rigorous methods for recording and testing instruments is best, but the major resonances are quite robust and identifiable.

Even though different types of strings are used the basic acoustical structure of the instrument remains the same. The point is what the soundbox does with the vibrations from the strings.

~OK

So, basically you're proving the point, when players know that strings make the difference, but the instrumentation doesn't see the difference.

Link to comment
Share on other sites

players know that strings make the difference, but the instrumentation doesn't see the difference.

In the same way a digital camera can remember an image perfectly, but the mind cannot.

Difference is that the mind forms an instant edited perception of an image, not a photo, and the camera forms an image, not a perception.

Same goes for sound, we form an impression of sound in our minds.

The mechanics of sound based on one given violin shape may be of some consequence to the maker player and buyer, if only one shape is used, but what if we use different shapes and thicknesses ? Surely then it's all so variable that the testing of the instruments amounts to somewhat of a confusing excercise.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.



×
×
  • Create New...