"timbre refers to the perceptual quality of sounds"
"timbre is that attribute of sensation in terms of which a
listener can judge that two sounds having the same loudness
and pitch are dissimilar"
"the timbre of an instrument is the type of sound it makes"
"everything that is not loudness, pitch, or spatial perception"
So often used, but often so ambiguously defined. Timbre like pitch is tied to human perception, thus it is inherently subjective in nature. The last definition above is the most correct. Timbre is defined in terms of what it is not. It is all the qualities left over after describing pitch and loudness. Timbre is often determined by the harmonic content and dynamic character of the sound; it is a psychoacoustic property. Scales to rate or distinguish timbre have been developed and use terms like sharp & dull,compact & scattered, bright & dark, hollow & full. Although these terms do have meaningful correlations to sound, they are still very subjective and do not offer a systematic method for timbre classification.
The speech counterpart of musical timbre is the phoneme, which often carries some of the same ambiguity as the term timbre. Merriam Webster defines it as "An abstract unit of language that is clearly distinguished from a set of similar sounds corresponding to it." So the phoneme like timbre involves distinguising "similar" sounds. However, the phoneme has been extensively studied and described by linguists, and offers a highly developed and systematic description of vocal sound. The English cardinal vowels are only a faction of the distinct set of phonemes that make up the English language.
It would be nice to have a similar system to describe musical timbre. In this project from music 255 at CCRMA, based on the ideas of Professor Jonathan Berger, I propose a system of musical timbre classification based on english vocal phonemes. Through the use of formant analysis, acoustic instrument sounds and combinations of instrument sounds can be classified on a vowel scale in a very analytical way. This provides a platform for concise timbre description and comparison. The fricative vocal sounds provide a system for describing attack, and non-tonal musical elements.
This system can be used as a tool for teaching orchestration. Orchestration is largely concerned with the timbral combination of instrumental and sometimes electronic sounds. The phoneme classification system provides a platform for scientific orchestrational analysis. Also, it offers teachers a well-defined method for evaluating and critiquing student work.
Joined: Aug 08, 2009 Posts: 1007 Location: Out scouting . . .
Posted: Fri Apr 14, 2017 3:05 pm Post subject:
Nothing in the wav file resembles the sound of human speech to me. I sort of get the idea that the author wanted to approximate speech (perhaps based on some sort of formant analysis) by reconstructing the result of analysis using a combination of orchestral timbres.
The academic in me would have referred the student to Doug Slocum and his project with the voder. Maybe with clarinets, oboes, English horns, etc. a score and various techniques of bending the notes an orchestra could be used to more closely approximate speech sounds.*
A full cymbal crash really doesn't work to simulate fricative sounds to my ear. It sounds like a cymbal crash. JMO.
Although I could recognize some speech in it I have to agree that it is not speech, but I also think that is beyond the point. I think the score to be an illustration mainly of the idea to make a mapping between timbres of instruments and the human voice.
I found this idea to be an interesting one. I think because I've had the thought at times that instruments sometimes, seem to be mimicking the human voice. Maybe mainly so for for the theremin which often reminds me to a singing human. But I've also had such associations with violin sounds.
I've also been experimenting at times with a simple speech synthesizer, and have found the output to sound like musical instruments at times. _________________ Jan
Joined: Aug 08, 2009 Posts: 1007 Location: Out scouting . . .
Posted: Fri Apr 14, 2017 4:16 pm Post subject:
Blue Hell wrote:
Although I could recognize some speech in it I have to agree that it is not speech, but I also think that is beyond the point.
You might be right, but based on the description of the methods used, it seemed to me that was a large part of the point. I mean, why bother with formant analysis? Like Prokofiev's Peter and the Worlf where musical themes and instruments represented certain characters, syllables and words could be represented (mapped) using any arbitrary characteristics/parameters. What I got out of the idea of the formant analysis was an attempt to mimic the sounds of speech with the sounds of musical instruments.
As I said, I could be wrong in what the intent was. Well, the intent was to get a good grade in a class, but beyond that I would have to take a closer reading of the projects purpose and methods.
What is the purpose of all this analysis?
I am not suggesting that every single acoustic instrument or combination of instruments need be analyzed in this way. I hope that from a generalized framework as in the chart below, which I am calling the "vowel space", each instrument class can be identified and the timbral mixture of particular instruments can be accurately predicted and described within this system.
The musical examples that follow take a different approach. Acoustic instrument sounds or their combinations are being selected to mimic vocal phonemes. So rather than describing acoustic timbre in terms of phonemes, the phonemes are being emulated by acoustic instruments. These experiments are offered as an attempt to verify the validity of a phoneme based system for timbre classiicfation. This system is under development and I encourage comments, questions, and feedback.
I have bolded the section that leads me to believe that I understood the a major point of the project correctly. The project designer wants to "verify the validity" of the system by showing mimetic correlations.
However, with your creativity and inquisitive mind, you might be seeing some other usefulness to the mapping beyond the goal(s) of the project. I respect and applaud that. And if so, I'd say you deserve the credit.
I understand your point Steve, it does not work very well to reverse the thing, and you were right that there are better methods for generating speech. Even with a single piano you'll get better speech...
As to creativity, I am not sure what to do with this (if anything), just found it to be an interesting way to look at timbre, and how musical instruments seem to be modeled after the human voice.
But then again .. when it works so well with a piano .. one should be able to do better with an orchestra as well .. but that'd be a lot more solenoids _________________ Jan
My personal view is that this concept is vital to electronic music where the timbre of a "note" can change over time in a way similar to how the frequency of the notes changes in a conventional melody. Early composers didn't have the instruments or orchestras the we have. Electronic musicians have carried on the Klangfarbenmelodie concept and taken it to new extremes.
In the extreme, a musical composition can be only one note that changes timbre as the composition evolves. (Examples of this are too numerous to mention.) Stretching the concept a bit, a sequence can be considered a single note; if the timbre varies over time, then it has musical interest related to Klangfarbenmelodie.
Long stretches of continuously repeated drum machine and/or sequences become boring (to me at least) very quickly unless there is some tambre change over time.
In electronic music, the concept of timbre has been expanded to include spacial localization (spacialization?). _________________ --Howard
my music and other stuff
Joined: Jan 18, 2005 Posts: 4146 Location: Sweden
Audio files: 371
G2 patch files: 100
Posted: Mon Jun 26, 2017 1:33 pm Post subject:
"Klangfärg" (basically the same word as "Klangfarbe", the Germans probably had it first) is also the common swedish translation of "timbre". It's a nice word; "färg" means colour and "klang" is a general word for a musical sound, mostly meaning something that comes from a struck instrument (like a bell), but can also be used more generally.
I think "colour" is a nice, easy to understand part of a metaphor for timbre, even though it's not the whole story. _________________ Antimon's Window @soundcloud@Flattrhome - you can't explain music
Joined: Aug 14, 2011 Posts: 423 Location: Griffith, Indiana, USA
Audio files: 148
Posted: Wed Jul 12, 2017 7:06 pm Post subject:
Blue Hell wrote:
I've had the thought at times that instruments sometimes, seem to be mimicking the human voice.
More than once in the many books I've read about classical music and orchestration, certain instruments were said to be particularly expressive because they approximated the human voice, which was apparently considered the ideal instrument. This was usually mentioned in cases of instrument improvements.
This made me wonder if orchestral instrument designers deliberately attempted to approach human voice characteristics. It could just as easily be that instrument designers might were trying to create instruments more reliably expressive than the human voice. _________________ AEJOTZ is pronounced "A-Jotz"
electronics = magic
free albums at http://aejotz.bandcamp.com
listen to genre-defying synthetic music at http://sat-5.com
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You cannot attach files in this forum You can download files in this forum
Please support our site. If you click through and buy from our affiliate partners, we earn a small commission.