I could be wrong, but I don't think it's quite the same thing. I believe the signal would be out of phase negating many of the effects of the filter. I would recommend using [biquad~] and in pd-extended there is a [notch] object which takes care of the coefficients. This sounds much cleaner and more notch-like to my ear than subtracting the filtered output. There is an explanation in Miller's book if you like unit circle math: http://msp.ucsd.edu/techniques/latest/book-html/node144.html
Message: 4 Date: Tue, 22 Apr 2014 01:59:07 -0300 From: Alexandre Torres Porres porres@gmail.com Subject: Re: [PD] WG: Inverse bandpass filter To: Ingo ingo@miamiwave.com Cc: pd-list pd-list@iem.at Message-ID: CAEAsFmhD0HanLmv9vutcSQZjkZY69i7wMeBqQ+20S2riWYaakw@mail.gmail.com Content-Type: text/plain; charset="utf-8"
isn't it just subtract the audio from the filtered output?
I guess you can get inverse freq response just by that
cheers
2014-04-18 17:21 GMT-03:00 Ingo ingo@miamiwave.com:
You could send the original signal in parallel and invert the phase by multiplying with -1. You might have to delay the original signal in case that the processed signal gets also delayed by one or more blocks.
Ingo
Von: pd-list-bounces@iem.at [mailto:pd-list-bounces@iem.at] Im Auftrag
von
AP Vague Gesendet: Freitag, 18. April 2014 18:49 An: pd-list@iem.at Betreff: [PD] Inverse bandpass filter
Is there a simple way to make [bp~] or [vcf~] have an inverse function?
To
filter out, rather than pass a changing frequency value. Is the easiest way to do this with a combination of [lop~] and [hip~]?
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Thank you for that link, that's awesome.
On Tue, Apr 22, 2014 at 1:37 PM, Robert Esler robert@urbanstew.org wrote:
I could be wrong, but I don't think it's quite the same thing. I believe the signal would be out of phase negating many of the effects of the filter. I would recommend using [biquad~] and in pd-extended there is a [notch] object which takes care of the coefficients. This sounds much cleaner and more notch-like to my ear than subtracting the filtered output.
There is an explanation in Miller's book if you like unit circle math: http://msp.ucsd.edu/techniques/latest/book-html/node144.html
Message: 4 Date: Tue, 22 Apr 2014 01:59:07 -0300 From: Alexandre Torres Porres porres@gmail.com Subject: Re: [PD] WG: Inverse bandpass filter To: Ingo ingo@miamiwave.com Cc: pd-list pd-list@iem.at Message-ID: CAEAsFmhD0HanLmv9vutcSQZjkZY69i7wMeBqQ+20S2riWYaakw@mail.gmail.com Content-Type: text/plain; charset="utf-8"
isn't it just subtract the audio from the filtered output?
I guess you can get inverse freq response just by that
cheers
2014-04-18 17:21 GMT-03:00 Ingo ingo@miamiwave.com:
You could send the original signal in parallel and invert the phase by multiplying with -1. You might have to delay the original signal in case that the processed signal gets also delayed by one or more blocks.
Ingo
Von: pd-list-bounces@iem.at [mailto:pd-list-bounces@iem.atpd-list-bounces@iem.at]
Im Auftrag von
AP Vague Gesendet: Freitag, 18. April 2014 18:49 An: pd-list@iem.at Betreff: [PD] Inverse bandpass filter
Is there a simple way to make [bp~] or [vcf~] have an inverse function?
To
filter out, rather than pass a changing frequency value. Is the easiest way to do this with a combination of [lop~] and [hip~]?
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
can;t remember where I saw about this, but check this link
http://msp.ucsd.edu/techniques/v0.11/book-html/node141.html
see the quote "An easy and practical way to remove the zero-frequency component from an audio signal is to use a one-pole low-pass filter to extract it, and then subtract the result from the signal. The resulting transfer function is one minus the transfer function of the low-pass filter:"
doesn't it agree with what I said?
cheers
2014-04-22 14:37 GMT-03:00 Robert Esler robert@urbanstew.org:
I could be wrong, but I don't think it's quite the same thing. I believe the signal would be out of phase negating many of the effects of the filter. I would recommend using [biquad~] and in pd-extended there is a [notch] object which takes care of the coefficients. This sounds much cleaner and more notch-like to my ear than subtracting the filtered output.
There is an explanation in Miller's book if you like unit circle math: http://msp.ucsd.edu/techniques/latest/book-html/node144.html
Message: 4 Date: Tue, 22 Apr 2014 01:59:07 -0300 From: Alexandre Torres Porres porres@gmail.com Subject: Re: [PD] WG: Inverse bandpass filter To: Ingo ingo@miamiwave.com Cc: pd-list pd-list@iem.at Message-ID: CAEAsFmhD0HanLmv9vutcSQZjkZY69i7wMeBqQ+20S2riWYaakw@mail.gmail.com Content-Type: text/plain; charset="utf-8"
isn't it just subtract the audio from the filtered output?
I guess you can get inverse freq response just by that
cheers
2014-04-18 17:21 GMT-03:00 Ingo ingo@miamiwave.com:
You could send the original signal in parallel and invert the phase by multiplying with -1. You might have to delay the original signal in case that the processed signal gets also delayed by one or more blocks.
Ingo
Von: pd-list-bounces@iem.at [mailto:pd-list-bounces@iem.atpd-list-bounces@iem.at]
Im Auftrag von
AP Vague Gesendet: Freitag, 18. April 2014 18:49 An: pd-list@iem.at Betreff: [PD] Inverse bandpass filter
Is there a simple way to make [bp~] or [vcf~] have an inverse function?
To
filter out, rather than pass a changing frequency value. Is the easiest way to do this with a combination of [lop~] and [hip~]?
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Though with DC you don't have the issue of phase. I'm not an expert in filter math, but I assume that by the time your filtered audio (assuming its not DC) gets subtracted by the [-~] object it is out of phase with the original signal. Moreover, I hear a distinct difference. Maybe I'm not conceiving your statement properly and perhaps this discussion has been about DC all along
Regards
From: Alexandre Torres Porres porres@gmail.com Date: Tuesday, April 22, 2014 3:51 PM To: GCC robert@urbanstew.org Cc: apvague@gmail.com, Ingo ingo@miamiwave.com, pd-lista puredata pd-list@iem.at Subject: Re: [PD] WG: Inverse bandpass filter
can;t remember where I saw about this, but check this link
http://msp.ucsd.edu/techniques/v0.11/book-html/node141.html
see the quote "An easy and practical way to remove the zero-frequency component from an audio signal is to use a one-pole low-pass filter to extract it, and then subtract the result from the signal. The resulting transfer function is one minus the transfer function of the low-pass filter:"
doesn't it agree with what I said?
cheers
2014-04-22 14:37 GMT-03:00 Robert Esler robert@urbanstew.org:
I could be wrong, but I don't think it's quite the same thing. I believe the signal would be out of phase negating many of the effects of the filter. I would recommend using [biquad~] and in pd-extended there is a [notch] object which takes care of the coefficients. This sounds much cleaner and more notch-like to my ear than subtracting the filtered output. There is an explanation in Miller's book if you like unit circle math: http://msp.ucsd.edu/techniques/latest/book-html/node144.html
Message: 4 Date: Tue, 22 Apr 2014 01:59:07 -0300 From: Alexandre Torres Porres porres@gmail.com Subject: Re: [PD] WG: Inverse bandpass filter To: Ingo ingo@miamiwave.com Cc: pd-list pd-list@iem.at Message-ID: CAEAsFmhD0HanLmv9vutcSQZjkZY69i7wMeBqQ+20S2riWYaakw@mail.gmail.com Content-Type: text/plain; charset="utf-8"
isn't it just subtract the audio from the filtered output?
I guess you can get inverse freq response just by that
cheers
2014-04-18 17:21 GMT-03:00 Ingo ingo@miamiwave.com:
You could send the original signal in parallel and invert the phase by multiplying with -1. You might have to delay the original signal in case that the processed signal gets also delayed by one or more blocks.
Ingo
Von: pd-list-bounces@iem.at [mailto:pd-list-bounces@iem.at] Im Auftrag
von
AP Vague Gesendet: Freitag, 18. April 2014 18:49 An: pd-list@iem.at Betreff: [PD] Inverse bandpass filter
Is there a simple way to make [bp~] or [vcf~] have an inverse function?
To
filter out, rather than pass a changing frequency value. Is the easiest way to do this with a combination of [lop~] and [hip~]?
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
could be, I know nothing really about it. But I think I've read something that stated so. And I also tried it and saw that you could inverse filters like that.
cheers
2014-04-22 21:06 GMT-03:00 Robert Esler robert@urbanstew.org:
Though with DC you don't have the issue of phase. I'm not an expert in filter math, but I assume that by the time your filtered audio (assuming its not DC) gets subtracted by the [-~] object it is out of phase with the original signal. Moreover, I hear a distinct difference. Maybe I'm not conceiving your statement properly and perhaps this discussion has been about DC all along…
Regards
From: Alexandre Torres Porres porres@gmail.com Date: Tuesday, April 22, 2014 3:51 PM To: GCC robert@urbanstew.org Cc: apvague@gmail.com, Ingo ingo@miamiwave.com, pd-lista puredata < pd-list@iem.at>
Subject: Re: [PD] WG: Inverse bandpass filter
can;t remember where I saw about this, but check this link
http://msp.ucsd.edu/techniques/v0.11/book-html/node141.html
see the quote "An easy and practical way to remove the zero-frequency component from an audio signal is to use a one-pole low-pass filter to extract it, and then subtract the result from the signal. The resulting transfer function is one minus the transfer function of the low-pass filter:"
doesn't it agree with what I said?
cheers
2014-04-22 14:37 GMT-03:00 Robert Esler robert@urbanstew.org:
I could be wrong, but I don't think it's quite the same thing. I believe the signal would be out of phase negating many of the effects of the filter. I would recommend using [biquad~] and in pd-extended there is a [notch] object which takes care of the coefficients. This sounds much cleaner and more notch-like to my ear than subtracting the filtered output.
There is an explanation in Miller's book if you like unit circle math: http://msp.ucsd.edu/techniques/latest/book-html/node144.html
Message: 4 Date: Tue, 22 Apr 2014 01:59:07 -0300 From: Alexandre Torres Porres porres@gmail.com Subject: Re: [PD] WG: Inverse bandpass filter To: Ingo ingo@miamiwave.com Cc: pd-list pd-list@iem.at Message-ID: CAEAsFmhD0HanLmv9vutcSQZjkZY69i7wMeBqQ+20S2riWYaakw@mail.gmail.com Content-Type: text/plain; charset="utf-8"
isn't it just subtract the audio from the filtered output?
I guess you can get inverse freq response just by that
cheers
2014-04-18 17:21 GMT-03:00 Ingo ingo@miamiwave.com:
You could send the original signal in parallel and invert the phase by multiplying with -1. You might have to delay the original signal in case that the processed signal gets also delayed by one or more blocks.
Ingo
Von: pd-list-bounces@iem.at [mailto:pd-list-bounces@iem.atpd-list-bounces@iem.at]
Im Auftrag von
AP Vague Gesendet: Freitag, 18. April 2014 18:49 An: pd-list@iem.at Betreff: [PD] Inverse bandpass filter
Is there a simple way to make [bp~] or [vcf~] have an inverse function?
To
filter out, rather than pass a changing frequency value. Is the easiest way to do this with a combination of [lop~] and [hip~]?
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
-
Alexandre Torres Porres -
AP Vague -
Robert Esler