spectral analysis-resynthesis program

Can Akkoc wrote:

I am looking for a simple software spectrum analyzer_synthesizer that can do spectral
analysis of monophonic sounds from "natural" musical instruments and simulate such
sounds by generating tones with the same spectral content and at user-picked
fundamental frequencies.

I need this for determining the accuracy and resolution of my measurement system.

Can,

The requirements you give for additive sythesis/resynthesis along with a sampler like
"play back at desired frequency" narrows things down a lot. I know of no commercial
product that does this easily but I'm sure that either Csound or Max can do it, as long
as you can live with the default lengths/parameters of the FFTs. Otheriwse, you're into
a general programming language which will have lots of overhead.

However - it strikes me that if the goal is to determine the accuracy of your pitch
trackign method, you may not need all of the capabilities you've asked for, in which
case there might be simpler ways of accomplishing the task... what exactly do you
have that you are trying to test?

--Bill Sethares

Seems like Akkoc is looking for a sampler--like my old EPS or the ARS that allow you to record tones from a "natural" instrument and then tweek the frequencies they're played back at into different scales maintaining the original sound "spectrum".

Bruce Kanzelmeyer

Bill Sethares <sethares@ece.wisc.edu> wrote:

Can Akkoc wrote:

I am looking for a simple software spectrum analyzer_synthesizer that can do spectral
analysis of monophonic sounds from "natural" musical instruments and simulate such
sounds by generating tones with the same spectral content and at user-picked
fundamental frequencies.

I need this for determining the accuracy and resolution of my measurement system.

Can,

The requirements you give for additive sythesis/resynthesis along with a sampler like
"play back at desired frequency" narrows things down a lot. I know of no commercial
product that does this easily but I'm sure that either Csound or Max can do it, as long
as you can live with the default lengths/parameters of the FFTs. Otheriwse, you're into
a general programming language which will have lots of overhead.

However - it strikes me that if the goal is to determine the accuracy of your pitch
trackign method, you may not need all of the capabilities you've asked for, in which
case there might be simpler ways of accomplishing the task... what exactly do you
have that you are trying to test?

--Bill Sethares

---------------------------------
Do you Yahoo!?
New Yahoo! Photos - easier uploading and sharing

Bill Sethares <sethares@ece.wisc.edu> wrote:
Can Akkoc wrote:

I am looking for a simple software spectrum analyzer_synthesizer that can do spectral
analysis of monophonic sounds from "natural" musical instruments and simulate such
sounds by generating tones with the same spectral content and at user-picked
fundamental frequencies.

I need this for determining the accuracy and resolution of my measurement system.

Can,

The requirements you give for additive sythesis/resynthesis along with a sampler like
"play back at desired frequency" narrows things down a lot. I know of no commercial
product that does this easily but I'm sure that either Csound or Max can do it, as long
as you can live with the default lengths/parameters of the FFTs. Otheriwse, you're into
a general programming language which will have lots of overhead.

However - it strikes me that if the goal is to determine the accuracy of your pitch
trackign method, you may not need all of the capabilities you've asked for, in which
case there might be simpler ways of accomplishing the task... what exactly do you
have that you are trying to test?

--Bill Sethares

Prof. Sethares,

I certainly will look into Csound and/or Max and try to work my way through this problem.

As for testing the accuracy and resolution of my pitch tracking system, I can do it easily using a simple (pure) wave form - sine, triangular, sqaure, etc - from a standard tone generator. However, my gut tells me "proper" response for a pure wave form does not necessarily mean the measurement system is just as accurate/resolute when it comes to capturing the fundamental frequency embedded in very complex wave forms coming from a natural instrument. Am I being overly concerned about this aspect of pitch tracking?

Thanks.

Yahoo! Groups SponsorADVERTISEMENT

You do not need web access to participate. You may subscribe through
email. Send an empty email to one of these addresses:
tuning-subscribe@yahoogroups.com - join the tuning group.
tuning-unsubscribe@yahoogroups.com - unsubscribe from the tuning group.
tuning-nomail@yahoogroups.com - put your email message delivery on hold for the tuning group.
tuning-digest@yahoogroups.com - change your subscription to daily digest mode.
tuning-normal@yahoogroups.com - change your subscription to individual emails.
tuning-help@yahoogroups.com - receive general help information.

Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

Can Akkoc

can193849@yahoo.com

---------------------------------
Do you Yahoo!?
New Yahoo! Photos - easier uploading and sharing

Bruce Kanzelmeyer <bkanzelmeyer@yahoo.com> wrote:Seems like Akkoc is looking for a sampler--like my old EPS or the ARS that allow you to record tones from a "natural" instrument and then tweek the frequencies they're played back at into different scales maintaining the original sound "spectrum".

Bruce Kanzelmeyer

Bill Sethares <sethares@ece.wisc.edu> wrote:

Can Akkoc wrote:

I am looking for a simple software spectrum analyzer_synthesizer that can do spectral
analysis of monophonic sounds from "natural" musical instruments and simulate such
sounds by generating tones with the same spectral content and at user-picked
fundamental frequencies.

I need this for determining the accuracy and resolution of my measurement system.

Can,

The requirements you give for additive sythesis/resynthesis along with a sampler like
"play back at desired frequency" narrows things down a lot. I know of no commercial
product that does this easily but I'm sure that either Csound or Max can do it, as long
as you can live with the default lengths/parameters of the FFTs. Otheriwse, you're into
a general programming language which will have lots of overhead.

However - it strikes me that if the goal is to determine the accuracy of your pitch
trackign method, you may not need all of the capabilities you've asked for, in which
case there might be simpler ways of accomplishing the task... what exactly do you
have that you are trying to test?

--Bill Sethares

You're right on the mark Bruce. Could you let me in on your "old EPS or the ARS"?

Thanks.

Can Akkoc

can193849@yahoo.com

---------------------------------
Do you Yahoo!?
New Yahoo! Photos - easier uploading and sharing

--- In tuning@yahoogroups.com, Can Akkoc <can193849@y...> wrote:

>However, my gut tells me "proper" response for a pure wave form does
>not necessarily mean the measurement system is just as
>accurate/resolute when it comes to capturing the fundamental
>frequency embedded in very complex wave forms coming from a natural
>instrument.

Your gut may be right -- human auditory pitch tracking is much more
precise for 'natural' instruments such as human voice, winds, brass,
and bowed strings than it is for pure sine waves. The harmonic
overtones in these 'natural' instruments give the human pitch
processor much more information from which to estimate the
fundamental frequency.

On Dec 8, 2003, at 3:16 PM, Can Akkoc wrote:
>
> As for testing the accuracy and resolution of my pitch tracking > system, I can do it easily using a simple (pure) wave form - sine, > triangular, sqaure, etc - from a standard tone generator. However, my > gut tells me "proper" response for a pure wave form does not > necessarily mean the measurement system is just as accurate/resolute > when it comes to capturing the fundamental frequency embedded in very > complex wave forms coming from a natural instrument. Am I being overly > concerned about this aspect of pitch tracking?

I don't understand how the resynthesis capability is going to help you
determine the accuracy of the fundamental.

What is the test that you are envisioning conducting?

--BIll Sethares

Bill Sethares <sethares@ece.wisc.edu> wrote:

On Dec 8, 2003, at 3:16 PM, Can Akkoc wrote:
>
> As for testing the accuracy and resolution of my pitch tracking
> system, I can do it easily using a simple (pure) wave form - sine,
> triangular, sqaure, etc - from a standard tone generator. However, my
> gut tells me "proper" response for a pure wave form does not
> necessarily mean the measurement system is just as accurate/resolute
> when it comes to capturing the fundamental frequency embedded in very
> complex wave forms coming from a natural instrument. Am I being overly
> concerned about this aspect of pitch tracking?

(1) I don't understand how the resynthesis capability is going to help you determine the accuracy of the fundamental.

(2) What is the test that you are envisioning conducting?

--BIll Sethares

Prof. Sethares,

(1) I might be missing the essense of your question here. I will try to respond anyway. Please alert me if this is a non-answer to your question.

When I synthesize a waveform with a pre-determined fundamental that simulates the harmonic content of the musical sound coming out of a Turkish instrument, I am hoping I will be able to determine how faithfully my pitch tracking system senses the fundamental frequency in a test sound.

(2) I am envisioning conducting two tests. One test will involve taking many readings from the same test sound and looking at the coefficient of variation in the data acquired. This procedure will be repeated at various pitch levels within the range of the instrument. I am hoping the spread in the raw data thus obtained will give me a measure of accuaracy of the system.

Another test will involve feeding two sounds into the system - one sound at a time - with fundamentals that are very close and narrowing down the difference between the fundamentals until the system can no longer detect the difference between the two test sounds. This test should give me a measure of the resolution of the system.

Can Akkoc

Can Akkoc

can193849@yahoo.com

---------------------------------
Do you Yahoo!?
New Yahoo! Photos - easier uploading and sharing