Separation of speech from multiple speakers speaking in different microphones connected to a mixer

>You are a jackass.  The first post wasn't mine. Pay attention to
>details.

You are correct that the first post was not yours, and for that I
apologize.  However, you did not attach any date to what you researched,
other than "i read a book a couple of years ago" and left it at that, not
even a title.  At least I mentioned a title for the book I refer to.  It is
rather recent and the author is heavily active in the area.  You further
made statements clearly indicating that you don't understand the problem
statement to begin with, while getting ignorant with me for no apparent
reason other than some author said ICA doesn't work.  Well, a lot of others
certainly disagree, and I'm more willing to believe them than some yoyo
that doesn't even know what the cocktail party problem is.  I'm not the
jackass here.  I was merely trying to be helpful on a subject that I have
been studying rather intensely of late.

Randy, plenty of ICA apps abound.  The Hyvarinen book actually does a few
DS-CDMA implementations (as well as other separation problems).  One is a
channel estimation solution, and another, using FastICA (by Bingham and
Hyvarinen) is an actual symbol detection solution.  Some good authors for
research are Cichocki, Tsatsanis, Wang and Poor and one of the original
blind detection authors, Verdu.  A starting point for some recent work I've
done on this subject was this: 

K.-H. Yap, L. Guan and J. Evans, “Blind adaptive detection for CDMA
systems based on regularized independent component analysis,” Proc. of
IEEE Global Telecommunications Conference, San Antonio, USA, vol. 1, pp.
249-253, Nov. 2001.

This paper uses the actual spreading sequences, which differs from the
fully-blind methods employing Lagrange multipliers (Tsatsanis has done
several detectors in this area).  There are many new authors taking on the
task very recently (not just the big ones over at Helsinki) including Tulay
Adali more recently.  She is the technical chair at the IEEE International
Workshop on Machine Learning for Signal Processing.  There should be some
good papers coming out of this workshop.

If you'd like more info, send me an email.  I'll be more than happy to
point you towards some of the research I've done.

Mark

markt wrote:
>> Dig deeper into what? RADC (not the current name) was trying to solve
>> this problem in the 1980's-early 90's. I followed their efforts at
>> that time.  Their efforts up through that period were ultimately
>> admitted to not perform well.  The last solution that I am aware of
>> threw out most of the processing of previous solutions, because a much
>> simpler method produced more useful (not good) results.
> 
> I gave you two references, both of which are more current than the
> 80s/90s.  Do a simple Google search on "cocktail party problem" and ICA and
> you'll get a bevy of hits.

Pick any research topic and you will get a flood of hits. You'll even 
find plenty that claim wonderful things. Do you know of a hit backed up 
with a wave file or two that sounds impressive?

>> I would expect that slow convergence of ICA would be a problem because
>> the acoustic environment (head orientation, etc) often changes pretty
>> quickly.
> 
> I've implemented ICA on Rayleigh fading channels for a DS-CDMA system not
> unlike that used in US PCS systems.  It is not the end-all, yet, but
> there's a lot of research in this area.  Do the Google search I mention
> above and the folks at Helsinki U of T are the first that pop up (not
> coincidentally, Hyvarinen is the author of the book).
> 
>> BTW what is being described here is NOT the well known "Cocktail
>> Party" problem. That problem has to do with two microphones (or ears)
>> that are not mixed together.  Also, most people don't realize how the
>> actual "Cocktail Party" problem differs from what actually happens at
>> a cocktail party. If you are interested in the problem you can figure
>> out the huge difference.
> 
> I disagree.  The cocktail party problem as typically devised is simply the
> mixing of multiple speakers in a room.  Using multiple receivers, e.g. two
> ears, is merely one method to separate the signals (by exploiting phase
> differences which provides position information).  How they get mixed is
> really immaterial (well, assuming some linear summation), what matters is
> their independence.
> 
> Try not to be so ignorant in your replies and maybe others will be
> helpful, too.  Remember, YOU are asking for help, not me.
> 
> Mark

Steve

On May 5, 5:47&#4294967295;pm, "markt" <tak...@pericle.com> wrote:
> >You are a jackass. &#4294967295;The first post wasn't mine. Pay attention to
> >details.
>
> You are correct that the first post was not yours, and for that I
> apologize. &#4294967295;However, you did not attach any date to what you researched,
> other than "i read a book a couple of years ago" and left it at that, not
> even a title. &#4294967295;At least I mentioned a title for the book I refer to. &#4294967295;It is
> rather recent and the author is heavily active in the area. &#4294967295;You further
> made statements clearly indicating that you don't understand the problem
> statement to begin with, while getting ignorant with me for no apparent
> reason other than some author said ICA doesn't work. &#4294967295;Well, a lot of others
> certainly disagree, and I'm more willing to believe them than some yoyo
> that doesn't even know what the cocktail party problem is. &#4294967295;I'm not the
> jackass here. &#4294967295;I was merely trying to be helpful on a subject that I have
> been studying rather intensely of late.
>
> Randy, plenty of ICA apps abound. &#4294967295;The Hyvarinen book actually does a few
> DS-CDMA implementations (as well as other separation problems). &#4294967295;One is a
> channel estimation solution, and another, using FastICA (by Bingham and
> Hyvarinen) is an actual symbol detection solution. &#4294967295;Some good authors for
> research are Cichocki, Tsatsanis, Wang and Poor and one of the original
> blind detection authors, Verdu. &#4294967295;A starting point for some recent work I've
> done on this subject was this:
>
> K.-H. Yap, L. Guan and J. Evans, &#4294967295;Blind adaptive detection for CDMA
> systems based on regularized independent component analysis,&#4294967295; Proc. of
> IEEE Global Telecommunications Conference, San Antonio, USA, vol. 1, pp.
> 249-253, Nov. 2001.
>
> This paper uses the actual spreading sequences, which differs from the
> fully-blind methods employing Lagrange multipliers (Tsatsanis has done
> several detectors in this area). &#4294967295;There are many new authors taking on the
> task very recently (not just the big ones over at Helsinki) including Tulay
> Adali more recently. &#4294967295;She is the technical chair at the IEEE International
> Workshop on Machine Learning for Signal Processing. &#4294967295;There should be some
> good papers coming out of this workshop.
>
> If you'd like more info, send me an email. &#4294967295;I'll be more than happy to
> point you towards some of the research I've done.
>
> Mark

Mark,

I do know that people call both the one channel multiple speaker
problem  (AKA cochannel speakers problem), and the two channel
multiple speaker problem, and higher numbers of channels multiple
speaker problem, &#4294967295;the cocktail party problem.&#4294967295;  Which of these do you
suppose has any relevance to what happens at a cocktail party? Wait,
don&#4294967295;t answer!  It really doesn&#4294967295;t matter in my response as you will
see.

I looked on Amazon to see if I recognized the cover of the book I
read.  Now, this is truly ironic, actually quite humorous (at least to
me), a truly remarkably coincidence. The book I read is your
recommended reference &#4294967295;Independent Component Analysis&#4294967295;, by Aapo
Hyvarinen, Juha Karhunen, and Erikki Oja.

Recalling my previous post, pull out a copy of the book and read the
following pages (you could alternately pull these up on Amazon as I
did by searching the book for &#4294967295;cocktail&#4294967295;):

Page 148 shows 3 combinations of 3 audio sources (effectively 3
sources and 3 microphones) and identifies it as the &#4294967295;cock-tail party
problem&#4294967295;. The previous page (page 147) identifies this as an example
of one of the &#4294967295;practical applications&#4294967295; of ICA .  The 3 audio sources
are indeed linearly summed but not with the same weights, so they
produce 3 different signals, not all just mixed into one signal.

Page 446 has a section entitled &#4294967295;24.2 Audio Separation&#4294967295;. This and the
next page discuss significant problems associated with applying ICA to
the &#4294967295;cock-tail party problem&#4294967295;, until we get to page 448 where your
reference says &#4294967295;Because of these complications, audio separation is a
largely unsolved problem.&#4294967295;  Wow!  Just as I'd remembered it. So page
446 contradicts page 147 regarding this being a "practical
application" of ICA.

Now you have been quite generous with your characterization of me as
&#4294967295;ignorant&#4294967295;. A common definition of "ignorant" is &#4294967295;lacking information
or knowledge&#4294967295;. Without a doubt you are &#4294967295;ignorant&#4294967295; of the contents of
your own recommended reference regarding the &#4294967295;cocktail party problem&#4294967295;.
Apparently my memory at my almost AARP age still functions better than
your own.  I read a book 3 or so years ago, didn&#4294967295;t remember the title
because the book was useless for my application of interest, and I can
still recall the details. I would guess you have used this revered
text much more often, much more recently, and definitely are much more
interested in the subject of ICA , possibly actually owning the book,
yet your memory fails you, making you &#4294967295;ignorant&#4294967295;.

Now you have objected that there ARE applications of ICA. I never said
there weren&#4294967295;t any. You say you have found one that works for you.
Great!  But, based on your recommended reference, the topic at hand
isn&#4294967295;t going to be solved with ICA.

You should be feeling pretty foolish by now (as well as &#4294967295;ignorant&#4294967295;),
so I will ask you a question that you may use to partially redeem
yourself.  On the author&#4294967295;s website they have a demonstration of ICA
applied to the &#4294967295;cocktail party problem&#4294967295;.  Now, clearly neither of us
know exactly what they did, but based on your extensive experience,
which should have developed a healthy dose of skepticism regarding
demonstrations, what MIGHT the authors have done to make an impressive
demonstration of this method, that would not be representative of a
real-life application?  When you get older you will get better at
identifying probable idealized demonstrations. After all, they say it
doesn't work practically, but show a demo of it working.

As for you statement &#4294967295;I was merely trying to be helpful on a subject
that I have been studying rather intensely of late&#4294967295;, you haven&#4294967295;t been
helpful to anyone to date, so study more intensely!  Answer the demo
question if you wish, it is a good question, and your observations
might be helpful to someone.

Please don&#4294967295;t embarrass yourself further.

Dirk

dbell wrote:
> On May 5, 5:47 pm, "markt" <tak...@pericle.com> wrote:
>>> You are a jackass.  The first post wasn't mine. Pay attention to
>>> details.
>> You are correct that the first post was not yours, and for that I
>> apologize.  However, you did not attach any date to what you researched,
>> other than "i read a book a couple of years ago" and left it at that, not
>> even a title.  At least I mentioned a title for the book I refer to.  It is
>> rather recent and the author is heavily active in the area.  You further
>> made statements clearly indicating that you don't understand the problem
>> statement to begin with, while getting ignorant with me for no apparent
>> reason other than some author said ICA doesn't work.  Well, a lot of others
>> certainly disagree, and I'm more willing to believe them than some yoyo
>> that doesn't even know what the cocktail party problem is.  I'm not the
>> jackass here.  I was merely trying to be helpful on a subject that I have
>> been studying rather intensely of late.
>>
>> Randy, plenty of ICA apps abound.  The Hyvarinen book actually does a few
>> DS-CDMA implementations (as well as other separation problems).  One is a
>> channel estimation solution, and another, using FastICA (by Bingham and
>> Hyvarinen) is an actual symbol detection solution.  Some good authors for
>> research are Cichocki, Tsatsanis, Wang and Poor and one of the original
>> blind detection authors, Verdu.  A starting point for some recent work I've
>> done on this subject was this:
>>
>> K.-H. Yap, L. Guan and J. Evans, &#4294967295;Blind adaptive detection for CDMA
>> systems based on regularized independent component analysis,&#4294967295; Proc. of
>> IEEE Global Telecommunications Conference, San Antonio, USA, vol. 1, pp.
>> 249-253, Nov. 2001.
>>
>> This paper uses the actual spreading sequences, which differs from the
>> fully-blind methods employing Lagrange multipliers (Tsatsanis has done
>> several detectors in this area).  There are many new authors taking on the
>> task very recently (not just the big ones over at Helsinki) including Tulay
>> Adali more recently.  She is the technical chair at the IEEE International
>> Workshop on Machine Learning for Signal Processing.  There should be some
>> good papers coming out of this workshop.
>>
>> If you'd like more info, send me an email.  I'll be more than happy to
>> point you towards some of the research I've done.
>>
>> Mark
> 
> Mark,
> 
> I do know that people call both the one channel multiple speaker
> problem  (AKA cochannel speakers problem), and the two channel
> multiple speaker problem, and higher numbers of channels multiple
> speaker problem, &#4294967295;the cocktail party problem.&#4294967295;  Which of these do you
> suppose has any relevance to what happens at a cocktail party? Wait,
> don&#4294967295;t answer!  It really doesn&#4294967295;t matter in my response as you will
> see.
> 
> I looked on Amazon to see if I recognized the cover of the book I
> read.  Now, this is truly ironic, actually quite humorous (at least to
> me), a truly remarkably coincidence. The book I read is your
> recommended reference &#4294967295;Independent Component Analysis&#4294967295;, by Aapo
> Hyvarinen, Juha Karhunen, and Erikki Oja.
> 
> Recalling my previous post, pull out a copy of the book and read the
> following pages (you could alternately pull these up on Amazon as I
> did by searching the book for &#4294967295;cocktail&#4294967295;):
> 
> Page 148 shows 3 combinations of 3 audio sources (effectively 3
> sources and 3 microphones) and identifies it as the &#4294967295;cock-tail party
> problem&#4294967295;. The previous page (page 147) identifies this as an example
> of one of the &#4294967295;practical applications&#4294967295; of ICA .  The 3 audio sources
> are indeed linearly summed but not with the same weights, so they
> produce 3 different signals, not all just mixed into one signal.
> 
> Page 446 has a section entitled &#4294967295;24.2 Audio Separation&#4294967295;. This and the
> next page discuss significant problems associated with applying ICA to
> the &#4294967295;cock-tail party problem&#4294967295;, until we get to page 448 where your
> reference says &#4294967295;Because of these complications, audio separation is a
> largely unsolved problem.&#4294967295;  Wow!  Just as I'd remembered it. So page
> 446 contradicts page 147 regarding this being a "practical
> application" of ICA.
> 
> Now you have been quite generous with your characterization of me as
> &#4294967295;ignorant&#4294967295;. A common definition of "ignorant" is &#4294967295;lacking information
> or knowledge&#4294967295;. Without a doubt you are &#4294967295;ignorant&#4294967295; of the contents of
> your own recommended reference regarding the &#4294967295;cocktail party problem&#4294967295;.
> Apparently my memory at my almost AARP age still functions better than
> your own.  I read a book 3 or so years ago, didn&#4294967295;t remember the title
> because the book was useless for my application of interest, and I can
> still recall the details. I would guess you have used this revered
> text much more often, much more recently, and definitely are much more
> interested in the subject of ICA , possibly actually owning the book,
> yet your memory fails you, making you &#4294967295;ignorant&#4294967295;.
> 
> Now you have objected that there ARE applications of ICA. I never said
> there weren&#4294967295;t any. You say you have found one that works for you.
> Great!  But, based on your recommended reference, the topic at hand
> isn&#4294967295;t going to be solved with ICA.
> 
> You should be feeling pretty foolish by now (as well as &#4294967295;ignorant&#4294967295;),
> so I will ask you a question that you may use to partially redeem
> yourself.  On the author&#4294967295;s website they have a demonstration of ICA
> applied to the &#4294967295;cocktail party problem&#4294967295;.  Now, clearly neither of us
> know exactly what they did, but based on your extensive experience,
> which should have developed a healthy dose of skepticism regarding
> demonstrations, what MIGHT the authors have done to make an impressive
> demonstration of this method, that would not be representative of a
> real-life application?  When you get older you will get better at
> identifying probable idealized demonstrations. After all, they say it
> doesn't work practically, but show a demo of it working.
> 
> As for you statement &#4294967295;I was merely trying to be helpful on a subject
> that I have been studying rather intensely of late&#4294967295;, you haven&#4294967295;t been
> helpful to anyone to date, so study more intensely!  Answer the demo
> question if you wish, it is a good question, and your observations
> might be helpful to someone.
> 
> Please don&#4294967295;t embarrass yourself further.
> 
> Dirk

Good old academic's bait and switch. Get them to pay for the material, 
then tell them at the end its useless. :-)

Steve

Steve Underwood wrote:
> dbell wrote:
>>
>> I do know that people call both the one channel multiple speaker
>> problem  (AKA cochannel speakers problem), and the two channel
>> multiple speaker problem, and higher numbers of channels multiple
>> speaker problem, �the cocktail party problem.�  Which of these do you
>> suppose has any relevance to what happens at a cocktail party? Wait,
>> don�t answer!  It really doesn�t matter in my response as you will
>> see.
>>
>> I looked on Amazon to see if I recognized the cover of the book I
>> read.  Now, this is truly ironic, actually quite humorous (at least to
>> me), a truly remarkably coincidence. The book I read is your
>> recommended reference �Independent Component Analysis�, by Aapo
>> Hyvarinen, Juha Karhunen, and Erikki Oja.
>>
>> Recalling my previous post, pull out a copy of the book and read the
>> following pages (you could alternately pull these up on Amazon as I
>> did by searching the book for �cocktail�):
>>
>> Page 148 shows 3 combinations of 3 audio sources (effectively 3
>> sources and 3 microphones) and identifies it as the �cock-tail party
>> problem�. The previous page (page 147) identifies this as an example
>> of one of the �practical applications� of ICA .  The 3 audio sources
>> are indeed linearly summed but not with the same weights, so they
>> produce 3 different signals, not all just mixed into one signal.
>>
>> Page 446 has a section entitled �24.2 Audio Separation�. This and the
>> next page discuss significant problems associated with applying ICA to
>> the �cock-tail party problem�, until we get to page 448 where your
>> reference says �Because of these complications, audio separation is a
>> largely unsolved problem.�  Wow!  Just as I'd remembered it. So page
>> 446 contradicts page 147 regarding this being a "practical
>> application" of ICA.
>>
>> Now you have been quite generous with your characterization of me as
>> �ignorant�. A common definition of "ignorant" is �lacking information
>> or knowledge�. Without a doubt you are �ignorant� of the contents of
>> your own recommended reference regarding the �cocktail party problem�.
>> Apparently my memory at my almost AARP age still functions better than
>> your own.  I read a book 3 or so years ago, didn�t remember the title
>> because the book was useless for my application of interest, and I can
>> still recall the details. I would guess you have used this revered
>> text much more often, much more recently, and definitely are much more
>> interested in the subject of ICA , possibly actually owning the book,
>> yet your memory fails you, making you �ignorant�.
>>
>> Now you have objected that there ARE applications of ICA. I never said
>> there weren�t any. You say you have found one that works for you.
>> Great!  But, based on your recommended reference, the topic at hand
>> isn�t going to be solved with ICA.
>>
>> You should be feeling pretty foolish by now (as well as �ignorant�),
>> so I will ask you a question that you may use to partially redeem
>> yourself.  On the author�s website they have a demonstration of ICA
>> applied to the �cocktail party problem�.  Now, clearly neither of us
>> know exactly what they did, but based on your extensive experience,
>> which should have developed a healthy dose of skepticism regarding
>> demonstrations, what MIGHT the authors have done to make an impressive
>> demonstration of this method, that would not be representative of a
>> real-life application?  When you get older you will get better at
>> identifying probable idealized demonstrations. After all, they say it
>> doesn't work practically, but show a demo of it working.
>>
>> As for you statement �I was merely trying to be helpful on a subject
>> that I have been studying rather intensely of late�, you haven�t been
>> helpful to anyone to date, so study more intensely!  Answer the demo
>> question if you wish, it is a good question, and your observations
>> might be helpful to someone.
>>
>> Please don�t embarrass yourself further.
>>
>> Dirk
> 
> Good old academic's bait and switch. Get them to pay for the material, 
> then tell them at the end its useless. :-)
> 
> Steve

On reflection, I shouldn't complain about that. At least they told you 
its useless. The really nasty academic trick is the paper that ends with 
a bit of hand waving that implies the remainder of the problem is 
trivial, when it was actually beyond the author. The skilled writer 
makes the triviality of those odds and ends seem tantilisingly 
plausible, and wastes hours of your time. :-)

Regards,
Steve

On May 6, 11:25&#4294967295;am, Steve Underwood <ste...@dis.org> wrote:
> Steve Underwood wrote:
> > dbell wrote:
>
> >> I do know that people call both the one channel multiple speaker
> >> problem &#4294967295;(AKA cochannel speakers problem), and the two channel
> >> multiple speaker problem, and higher numbers of channels multiple
> >> speaker problem, &#4294967295;the cocktail party problem.&#4294967295; &#4294967295;Which of these do you
> >> suppose has any relevance to what happens at a cocktail party? Wait,
> >> don&#4294967295;t answer! &#4294967295;It really doesn&#4294967295;t matter in my response as you will
> >> see.
>
> >> I looked on Amazon to see if I recognized the cover of the book I
> >> read. &#4294967295;Now, this is truly ironic, actually quite humorous (at least to
> >> me), a truly remarkably coincidence. The book I read is your
> >> recommended reference &#4294967295;Independent Component Analysis&#4294967295;, by Aapo
> >> Hyvarinen, Juha Karhunen, and Erikki Oja.
>
> >> Recalling my previous post, pull out a copy of the book and read the
> >> following pages (you could alternately pull these up on Amazon as I
> >> did by searching the book for &#4294967295;cocktail&#4294967295;):
>
> >> Page 148 shows 3 combinations of 3 audio sources (effectively 3
> >> sources and 3 microphones) and identifies it as the &#4294967295;cock-tail party
> >> problem&#4294967295;. The previous page (page 147) identifies this as an example
> >> of one of the &#4294967295;practical applications&#4294967295; of ICA . &#4294967295;The 3 audio sources
> >> are indeed linearly summed but not with the same weights, so they
> >> produce 3 different signals, not all just mixed into one signal.
>
> >> Page 446 has a section entitled &#4294967295;24.2 Audio Separation&#4294967295;. This and the
> >> next page discuss significant problems associated with applying ICA to
> >> the &#4294967295;cock-tail party problem&#4294967295;, until we get to page 448 where your
> >> reference says &#4294967295;Because of these complications, audio separation is a
> >> largely unsolved problem.&#4294967295; &#4294967295;Wow! &#4294967295;Just as I'd remembered it. So page
> >> 446 contradicts page 147 regarding this being a "practical
> >> application" of ICA.
>
> >> Now you have been quite generous with your characterization of me as
> >> &#4294967295;ignorant&#4294967295;. A common definition of "ignorant" is &#4294967295;lacking information
> >> or knowledge&#4294967295;. Without a doubt you are &#4294967295;ignorant&#4294967295; of the contents of
> >> your own recommended reference regarding the &#4294967295;cocktail party problem&#4294967295;.
> >> Apparently my memory at my almost AARP age still functions better than
> >> your own. &#4294967295;I read a book 3 or so years ago, didn&#4294967295;t remember the title
> >> because the book was useless for my application of interest, and I can
> >> still recall the details. I would guess you have used this revered
> >> text much more often, much more recently, and definitely are much more
> >> interested in the subject of ICA , possibly actually owning the book,
> >> yet your memory fails you, making you &#4294967295;ignorant&#4294967295;.
>
> >> Now you have objected that there ARE applications of ICA. I never said
> >> there weren&#4294967295;t any. You say you have found one that works for you.
> >> Great! &#4294967295;But, based on your recommended reference, the topic at hand
> >> isn&#4294967295;t going to be solved with ICA.
>
> >> You should be feeling pretty foolish by now (as well as &#4294967295;ignorant&#4294967295;),
> >> so I will ask you a question that you may use to partially redeem
> >> yourself. &#4294967295;On the author&#4294967295;s website they have a demonstration of ICA
> >> applied to the &#4294967295;cocktail party problem&#4294967295;. &#4294967295;Now, clearly neither of us
> >> know exactly what they did, but based on your extensive experience,
> >> which should have developed a healthy dose of skepticism regarding
> >> demonstrations, what MIGHT the authors have done to make an impressive
> >> demonstration of this method, that would not be representative of a
> >> real-life application? &#4294967295;When you get older you will get better at
> >> identifying probable idealized demonstrations. After all, they say it
> >> doesn't work practically, but show a demo of it working.
>
> >> As for you statement &#4294967295;I was merely trying to be helpful on a subject
> >> that I have been studying rather intensely of late&#4294967295;, you haven&#4294967295;t been
> >> helpful to anyone to date, so study more intensely! &#4294967295;Answer the demo
> >> question if you wish, it is a good question, and your observations
> >> might be helpful to someone.
>
> >> Please don&#4294967295;t embarrass yourself further.
>
> >> Dirk
>
> > Good old academic's bait and switch. Get them to pay for the material,
> > then tell them at the end its useless. :-)
>
> > Steve
>
> On reflection, I shouldn't complain about that. At least they told you
> its useless. The really nasty academic trick is the paper that ends with
> a bit of hand waving that implies the remainder of the problem is
> trivial, when it was actually beyond the author. The skilled writer
> makes the triviality of those odds and ends seem tantilisingly
> plausible, and wastes hours of your time. :-)
>
> Regards,
> Steve- Hide quoted text -
>
> - Show quoted text -

What I have always been amazed at was the ICASSP (is it still called
that?) presentations where there were huge numbers of alternative,
optimal, .... presentations, several years in a row, that solve the
same half of a problem that no one has a clue how to solve the other
half.

Dirk

Steve Underwood wrote:

> On reflection, I shouldn't complain about that. At least they told you 
> its useless. The really nasty academic trick is the paper that ends with 
> a bit of hand waving that implies the remainder of the problem is 
> trivial, when it was actually beyond the author. The skilled writer 
> makes the triviality of those odds and ends seem tantilisingly 
> plausible, and wastes hours of your time. :-)

I have heard this method put to good use by math students
with smart TAs.

-- glen

On 6 Mai, 17:25, Steve Underwood <ste...@dis.org> wrote:
> Steve Underwood wrote:

> > Good old academic's bait and switch. Get them to pay for the material,
> > then tell them at the end its useless. :-)
>
> > Steve
>
> On reflection, I shouldn't complain about that. At least they told you
> its useless. The really nasty academic trick is the paper that ends with
> a bit of hand waving that implies the remainder of the problem is
> trivial, when it was actually beyond the author. The skilled writer
> makes the triviality of those odds and ends seem tantilisingly
> plausible, and wastes hours of your time. :-)

I have mentioned this one a few times before:

Soares, Siderius and Jesus: "Source localization in a time-varying
     ocean waveguide", Journ. Ac. Soc. Am., V 112 no 5, p 1879, 2002

The problem deals with passive sonar, i.e. listening devices
where the one tries to find the position of the sound source,
as well as detect it, with as little effort as possible. The
general idea is to deploy the devices and then leave them
alone and preferably undetected.

These guys set up a huge experiment, with a sensor array and a
sound source 10 or 12 km apart, at sea. The sensors record the
sound, and a huge effort is spent on computing the relative
source position (which is known, since ths is a controlled
experiment).

This particular experiment was unique at the time, as they
deployed a vessel running back and forth with a chain of
oceangraphic sensors. Basically, they scanned the relevant
parameters as completely and in as real-time as practically
possible at sea.

The interesting part was that the ocenography changed very fast.
In the matter of minutes. Measurements made by the extra vessel
on one pass were significantly different from the measurements
made on the next. And they were not able to deduce the relative
source position unless they used the very latest oceanographic
data - data that were measured literally simultaneous with the
recording of the sound to be processed.

So the authors conclude with that 'this method for sound
localization is robust for the purpose.' The fact that a
continuous measurement of the oceanography between the
covert sensor array and the unknown (and equally covert)
sound source was required for the processing to work,
escaped the authors completely.

I am sure that paper looked very well on the reference
list accompanying their next application for project
funding.

Rune

On 2 &#4294967295;&#4294967295;&#4294967295;, 20:00, "markt" <tak...@pericle.com> wrote:
> This is the well-known "cocktail party" problem. &#4294967295;

Sorry for coming back to initial problem.
Can someone give me an answer please?
How many voices or instruments that are singing or playing at one same
moment can some ordinary person distinguish? Assume that person just
listens the recorded mono file and says here are N persons in a choir
or here are M different instruments playing. A person is, say, an
ordinary musician, not Mozart. Also let us assume each voice or each
instrument may play an individual note.
Thanks in advance.
Vladimir.

>I do know that people call both the one channel multiple speaker
>problem  (AKA cochannel speakers problem), and the two channel
>multiple speaker problem, and higher numbers of channels multiple
>speaker problem, =93the cocktail party problem.=94  Which of these do
you
>suppose has any relevance to what happens at a cocktail party? Wait,
>don=92t answer!  It really doesn=92t matter in my response as you will
>see.

>The 3 audio sources
>are indeed linearly summed but not with the same weights, so they
>produce 3 different signals, not all just mixed into one signal.
>

Big deal, whether they are scaled or not.  LINEAR means sum plus scale. 
This is a basic concept in linear algebra.  That has no bearing on the
cocktail party problem.  The goal of ICA, or any linear separation problem,
is to estimate the mixing matrix, scales included.  Most ICA solutions have
a scale ambiguity anyway (though there are ways to work around any phase
ambiguity).

>Because of these complications, audio separation is a
>largely unsolved problem.=94  Wow!  Just as I'd remembered it. So page
>446 contradicts page 147 regarding this being a "practical
>application" of ICA.
>

Um, gee, the book was written in 2001, and it is now 2008, hence my
comment "dig deeper."  I never said the book solved the problem.

>Now you have been quite generous with your characterization of me as
>=93ignorant=94. A common definition of "ignorant" is =93lacking
information
>or knowledge=94.

Yes, clearly.  You ignorantly assumed that I said the book solved the
problem, and you STILL don't understand the difference between a problem
statement and a potential solution, as evidenced by this silly post.  You
also put the book down and assumed that since it had not been solved, it
could not be solved and thus "ICA is no good."  Good detective work there.

>Without a doubt you are =93ignorant=94 of the contents of
>your own recommended reference regarding the =93cocktail party
problem=94.
>Apparently my memory at my almost AARP age still functions better than
>your own.

I fully know the contents of the book, and not once did I ever say that
the book solved the problem.  You did, not me.  I simply provided that as a
reference.

> rant

You're still ignorant.

Mark