Re: [LAD] New release of jconv

[alex stone]

On Wed, Mar 11, 2009 at 2:59 AM, Fons Adriaensen wrote:

> On Wed, Mar 11, 2009 at 12:45:17AM +0100, Jörn Nettingsmeier wrote:
>
> > wow. thanks for the new example configurations and the extensive
> > comments! very helpful.
>
> I hope you like the AMB ones...
>
> > the convolver itself is a bit of a marketing disaster, though. i mean,
> > uh, it just convolves. No fancy new features, such as UltraLowJitter, or
> > at least TrueMultiplyAndAdd (for purists, as opposed to FFT-based, for
> > that extra dash of depth and clarity). And TubeMode would be cool as
> > well. but one cannot have it all, i assume. ;)
>
> :-) I'll take the IR of a long tube one if these days.
>
> > in the reverb configs, you mention that you cut off the direct sound by
> > setting an offset, which will then make the convolutions suitable to use
> > in a classic "aux send" setup, so that the reverb return will be 100%
> wet.
> > but why do you remove the first 5 ms of the room response?
>
> Because it's not room response but speaker response
> (mainly LF, if the source was not well equalised),
> or reflections from very close objects which are
> not perceived as reflections but just add a nasty
> colouration.
>
> Ciao,
>
> --
> FA
>
> Laboratorio di Acustica ed Elettroacustica
> Parma, Italia
>
> Be quiet, Master Land; and you, Professor,
> will you be so good as to listen to me ?
> ___
> Linux-audio-dev mailing list
> Linux-audio-dev@lists.linuxaudio.org
> http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev
>


Fons,

Thanks for the update. I was up early this morning testing it, and all goes
fine.
The new AMB configs are cool, and the more extensive user friendly comments
will certainly help.

Alex.



-- 
Parchment Studios (It started as a joke...)
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] Strange network problem ?

[Bengt Gördén]

Den Wednesday 11 March 2009 00.40.44 skrev Fons Adriaensen:
> Hello all,
>
> Today I had a strange problem when using the
> WFS system here.
>
> The 'master' PC sends a multicast message every
> 1024 samples (21.333 ms) to all 'render' PCs.
> This has to arrive on time, and when it's late
> the renderers will mute their output and report
> the error in their status messages.
>
> Today I used some ssh -X logins from my laptop
> in the WFS room to the WFS master to run Ardour
> and some other apps for a demo. All this worked
> well all the time, as it has done before.
>
> I left everything running when going for lunch
> with our visitors, and when I returned restarted
> Ardour to listen again. I got a lots 'of 'message
> too late' errors from the rendering machines, and
> interrupted sound. Strange enough this seemed to
> be related to the _volume_ of the sounds...
>
> In other words to Ardour's level meters.
> Restarting Ardour in a new ssh login did not
> help, but running it directly on the WFS master
> solved the problem. So apparently the network
> traffic required to update Ardour's meters was
> causing the delays. And clearly the whole remote
> X session was slower than normal. CPU loads
> looked normal.
>
> Now all this should be peanuts for a Gbit
> network that has no other traffic at all,
> and it worked perfectly before. I just never
> left it running for such a long time.
>
> Anyone an idea as to what is happening here,
> and how it could be cured ?

I can't say that I can help you but 19 years as a network engineer can come in 
handy.

The most obvious is that the network is overloaded in some way. Doesn't need 
to be in terms of bandwidth. It can be things like spanning-tree going 
heywire in switches. Redirects is another thing that comes to my mind. A 
sophisticated (there are things like slow pings of IPv6 addresses that fills 
up the table in a switch) DoS might also be the case.

If we assume it's the network that causes the problem you obviously have the 
switches and the routers that can cause the problems not to say firewalls. 
What network set up (equipment and such) do you have?

Is the network protected in some way? 

Have you made some performance tests? iperf is handy.

TCP:
server:> iperf -s
client:> iperf -c server

UDP:
server:> iperf -su
client:> iperf -c server -u


/bengan
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] Speech noise removal

[Olivier Guilyardi]

Julien Claassen wrote:
> Hi!
>   Csound also has an API, tat might help there. You can use it in C.
>   But I like the idea of a plugin. If you do something, it might be nice
> to prvide for usage in a text-based environment as well. A plugin would
> be good there.

I'll see what I can do about that, I can't promise anything though.

--
  Olivier
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] Speech noise removal

[Olivier Guilyardi]

alex stone wrote:
> 
> On Mon, Mar 9, 2009 at 2:59 PM, Olivier Guilyardi  > wrote:
> 
> alex stone wrote:
> 
> > If you're intent on automating a speech analysis, voice noise removal
> > device of some sort, then you might do well to start with a 'pre and
> > post' framework. Things like lipsmacking, glottal and nasal noise for
> > the end of phrase, etc, are fairly easy to identify, and generally
> occur
> > pre and post. So that may well be a decent percentage of any cleanup
> > done quickly. (Dependent of course on language. Cleaning up russian
> > would be a different 'module' to cleaning up French, or Finnish.)
> 
> That sounds encouraging. What to you mean by "pre and post" (sorry
> if that's an
> obvious question to you)?
> 
[...]
> Pre and post meaning the start and finish of a recorded wav or region.
> Example being the first few, and the last few, milliseconds or so. Most
> of this would be obvious to the ear, so i can imagine a means to edit
> this could be mechanised in some way. (Being
> careful, of course, not to dehumanise the original recording too far.)

Alright, got that. We've done some experimentations. On a 3 minutes of speech
recording, we got 53 noises, with the following repartition:

inspire: 37
expiration:   2
lips: 6
nose: 3
glottal:  1
inspire+lips: 4

That means "inspiring" (breathing in, between two phrases or groups of words)
noises makes 70% of the noises. These can be silenced, no need for frequency
filtering, because they always happen "pre and post" as you say, and they're
apparently always preceded and followed by small silences.

Here's the spectrogram+waveform of two inspire noises (the cursor, a white
vertical line, is on the noise). On each view appears 1 noise, surrounded by 
speech:

http://www.samalyse.com/code/speechfilter/inspire1.png
http://www.samalyse.com/code/speechfilter/inspire2.png

We've also measured the duration of 14 inspire noises. Except for 1, all of them
are under 1 second. The durations range from from 256ms to 1024ms, with an
average of 529ms.

An automatic way of removing these inspire noises may largely satisfy the users
I'm dealing with. Saving 70% of manual editing time is everything but marginal.
So I'm going to concentrate on that at first, leaving all lips, nose, ... noises
for later.

> Thinking further about modules, you might consider the inclusion (should
> you try this) of a user definable module, in which the user could set
> parameters. Consider the lone singer at home, or the voice over artist
> who uses the same 'voice' on a regular basis. They would tend to form
> phrases, and speech, in the same way, more often than not, including
> mouth noise, nasal, etc (Big generalisation here, but to get the
> point across...)
> If the user can use his or her own template each time as a start point,
> then it might prove more efficient, and definable, as a mechanised process.
> (Alex singing module, Olivier talking module, etc)

Correct me if I'm wrong, but if I code this as a plugin which exposes
parameters, I think that presets should be handled by the host, not the plugin.

Anyway, I'm not sure I could code this as a plugin, because, for detecting the
inspire noise, I would need to buffer something like 2 seconds of signal. I
suppose that might not be such a problem though, there's already plenty of non
RT-capable plugins...

Plus, before removal, visual/auditive review of the detected noises (in some
sort of audio editor) sounds quite important: there's alway a risk to confuse a
noise with the end of a phrase or an other element of speech. So I might need to
craft a little gui, or manage to integrate this detection into rezound, ardour,
etc..

Anyway, before this happens I need to find a way to detect the noises. A
colleague has told me that the best technology in this field currently involves
using a database of noise recordings. You then try to find these noises in the
signal by doing a more-or-less tolerant comparison in frequency domain.

However, looking at the above spectrograms and waveforms, I think there could be
a more algorithmic way of detecting these noises, thus avoiding the need for
such a database, given the following facts:

1 - on the waveform: their amplitude is much lower than speech
2 - on the spectrogram: the frequencies in the noise seem to spread rather
homogeneously (maybe a bit like white noise) where the speech contain noticeable
peaks under 1000Hz or so.

Do you think I could use these characteristics to detect the noises?

PS: Alex, maybe that you could try and improve the way you handle citations when
posting? That's not essential, but would make your replies more readable...

--
  Olivier





___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] Speech noise removal

[Julien Claassen]

Very kind of you! I'll just be looking forward to it... And if nothing comes 
round with the first version. That's surely an application up for longer 
parenting. :-)
   Kindest regards
Julien


Music was my first love and it will be my last (John Miles)

 FIND MY WEB-PROJECT AT: 
http://ltsb.sourceforge.net
the Linux TextBased Studio guide
=== AND MY PERSONAL PAGES AT: ===
http://www.juliencoder.de
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] Speech noise removal

[Chris Cannam]

On Wed, Mar 11, 2009 at 4:15 PM, Olivier Guilyardi  wrote:
> Plus, before removal, visual/auditive review of the detected noises (in some
> sort of audio editor) sounds quite important: there's alway a risk to confuse 
> a
> noise with the end of a phrase or an other element of speech. So I might need 
> to
> craft a little gui, or manage to integrate this detection into rezound, 
> ardour,
> etc..

I'm not sure I have any useful thoughts about the actual detection
method, but I notice your screenshots use SV.  If you could make a
Vamp plugin (turning, say, frequency-domain input into output features
with duration that identify the noisy regions), you would then be able
to review the results in either SV or Audacity.

And although you can't edit the audio in SV, by chance I'm just
planning to make a little command-line program to split audio files
according to the feature locations returned by Vamp plugins run on
them, which might help to provide one of the missing pieces.

Just a thought anyway.


Chris
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] Speech noise removal

[Olivier Guilyardi]

Julien Claassen wrote:
> Very kind of you! I'll just be looking forward to it... And if nothing
> comes round with the first version. That's surely an application up for
> longer parenting. :-)

After thinking about it, I think a plugin might be the way the go.. Not yet sure
about it though. Btw, what are the available text-based hosts for LV2 plugins?
Does ecasound support them?

Julien, by curiosity, what would you use this speech filter for?

--
  Olivier

___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] Speech noise removal

[Julien Claassen]

Salut Olivier!
  I don't think ecasound does it yet. Except if they are compatible to ladspa. 
I'll have to check. Besides that, there would be a simple test-host like 
applyplugin for LADSPA.
   I'd use this for my own speech recordings. Readings, perhaps testing it on 
singing as well. I once cut a film commentary, and I would have enjoyed such a 
tool tremendously. Because listening to it in ecasound, always stepping back 
and forth and fixing the right positions to ignore, is... Hard and 
frustrating. :-(
   About the RT-capability. There a plugin I like to use sometimes: vlevel. It 
needs up to 4 seconds. It's a look-ahead volume fader automation, so the 
signal is always a maximum level. I just have to remove this amount of time at 
the beginning of the resulting audio. It's quite OK and works nicely with 
ecasound.
   Kindest regards
Julien


Music was my first love and it will be my last (John Miles)

 FIND MY WEB-PROJECT AT: 
http://ltsb.sourceforge.net
the Linux TextBased Studio guide
=== AND MY PERSONAL PAGES AT: ===
http://www.juliencoder.de
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] Strange network problem ?

[Fons Adriaensen]

On Wed, Mar 11, 2009 at 11:35:15AM +0100, Bengt Gördén wrote:

> > Anyone an idea as to what is happening here,
> > and how it could be cured ?
> 
> I can't say that I can help you but 19 years as a network engineer can come 
> in 
> handy.
> 
> The most obvious is that the network is overloaded in some way. Doesn't need 
> to be in terms of bandwidth. It can be things like spanning-tree going 
> heywire in switches. Redirects is another thing that comes to my mind. A 
> sophisticated (there are things like slow pings of IPv6 addresses that fills 
> up the table in a switch) DoS might also be the case.
> 
> If we assume it's the network that causes the problem you obviously have the 
> switches and the routers that can cause the problems not to say firewalls. 
> What network set up (equipment and such) do you have?
> 
> Is the network protected in some way? 

I'll make a complete list of all the equipment
next monday (I don't return there before).

The network is completely isolated, there are just
the four WFS computers (and a fifth which was not
being used), and my Thinkpad R51.
Apart from the ssh -X connection the only traffic
are the messages I mentioned before (around 50/s)
and the status reports from the slave computers,
around 60/s for all three added. All messages are
less than an MTU.

> Have you made some performance tests? iperf is handy.
> 
> TCP:
> server:> iperf -s
> client:> iperf -c server
> 
> UDP:
> server:> iperf -su
> client:> iperf -c server -u

Good tip, will do.

Now you're here you may be able to comment on
another problem: the network adapter in the
master does not receive the multicast messages
it is sending (even if the TX socket has the
IP_MULTICAST_LOOP option set). Chip is Intel
85something, driver e1000e. 

Many thanks,

-- 
FA

Laboratorio di Acustica ed Elettroacustica
Parma, Italia

Be quiet, Master Land; and you, Professor,
will you be so good as to listen to me ?
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] New release of jconv

[Steve Fosdick]

On Tue, 2009-03-10 at 23:28 +0100, Fons Adriaensen wrote:
> Hello all,
> 
> Jconv-0.8.0 is now available at the usual place

Fons,

This appears to use a version of libsndfile that has ambisonic
functionality.  Is this in the mainstream libsndfile or is there a
forked or patched version somewhere?

Regards,
Steve.
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] New release of jconv

[Jörn Nettingsmeier]

Steve Fosdick wrote:
> On Tue, 2009-03-10 at 23:28 +0100, Fons Adriaensen wrote:
>> Hello all,
>>
>> Jconv-0.8.0 is now available at the usual place
> 
> Fons,
> 
> This appears to use a version of libsndfile that has ambisonic
> functionality.  Is this in the mainstream libsndfile or is there a
> forked or patched version somewhere?

you can edit the makefile to make it work without. you could also get a
pre-release from eric's website: http://www.mega-nerd.com/tmp/

but unless i'm very much mistaken, the current release does include .amb
support already.


___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] New release of jconv

[Fons Adriaensen]

On Wed, Mar 11, 2009 at 10:29:49PM +, Steve Fosdick wrote:

> > Jconv-0.8.0 is now available at the usual place
> 
> Fons,
> 
> This appears to use a version of libsndfile that has ambisonic
> functionality.  Is this in the mainstream libsndfile or is there a
> forked or patched version somewhere?

You can use the official release, see the Makefile
where you need to uncomment one line. The standard
libsndfile can read files with the ambisonic GUID,
so this will not affect jconv. It does affect some
unused (by jconv) functionality in the impdata class
which jconv shares with other apps.

Version 18 of libsndfile has been in pre-release
for something like two years now. You can find
it on Eric's site. If you decide to use it make
sure to take a recent one, as the AMB stuff was
broken in some older pre-release versions.

Ciao,

-- 
FA

Laboratorio di Acustica ed Elettroacustica
Parma, Italia

Be quiet, Master Land; and you, Professor,
will you be so good as to listen to me ?
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

[LAD] [ann] QuteCsound 0.4

[Andres Cabrera]

Hi all,

(And apologies for cross posting)

I'm pleased to announce QuteCsound version 0.4. This version
incorporates new features and bug fixes arising from the 3 previous
Release Candidate releases.

QuteCsound is a simple frontend for Csound featuring a highlighting
editor with autocomplete, interactive widgets and integrated help. It
is a cross-platform drop in replacement for MacCsound (it can open
files created in MacCsound), and aims to be a simple yet powerful and
complete development environment for Csound.

QuteCsound 0.4 has been tested on Windows, OS X, Linux and Solaris,
and it is free software released under the GPL.

You can get it here:
http://sourceforge.net/project/showfiles.php?group_id=227265&package_id=275156&release_id=664354

There's more information here:
http://qutecsound.sourceforge.net/

Questions, comments and suggestions can be posted to the main Csound
mailing list, or join the QuteCsound users mailing list at:
http://lists.sourceforge.net/lists/listinfo/qutecsound-users

Thanks very much to all the people (particularly Joachim Heintz), who
have provided great feedback and ideas to improve QuteCsound. Also a
big thanks to Andy Fillebrown for the fixes for Windows and for
preparing the Windows installers.

Enjoy!
Andrés


Changes from version 0.4RC2:

New:
- Added "_Browse" channel for value buttons, which open a browse file
dialog and send the selected file name to any widget with the same
channel which can receive strings, like LineEdit. Added this feature
to the reserved channels example.
- New tutorials and example files, with clearer names. The examples
can now be run directly WITHOUT saving, which is very handy especially
for testing audio IO.
- Console text font color can now be changed
- Added a close tab button on the right corner of the tabs panel.
- Modified execute code for Windows (for external utilities and terminal)
- Added option to enable/disable line wrap
- Added clear option to console context menus
- MacCsound output filename is now parsed and added to the options
- Added new FM and additive examples

Fixed:
- Improved stability
- Files with extensions different to csd, orc and sco can now be opened
- External programs and terminal are now launched correctly on all platforms
- Fixed storing changes in widget panel when tab changes
- Graph widget now displays graphs generated by ftgen
- Widget values are now properly sent for the first control period
- Files now open properly on OS X when double clicked from the Finder
- Fixed problem when buffering messages
- Fixed problems with filenames with spaces when rendering or opening
in external programs
- MacCsound sections are only created if save Widgets preference is set
- New default file more consistent with current state.
- Files are now not marked as modified when changing tabs
- Fixed path of Manual for newer versions
- Fixed resizing and moving widget edit frames when changed from the
properties dialog
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] New release of jconv

[Paul Davis]

On Wed, Mar 11, 2009 at 6:52 PM, Fons Adriaensen wrote:

>
> Version 18 of libsndfile has been in pre-release
> for something like two years now.


actually, 18 has been out for a few months, and now 19 has been "let go" as
well.
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] New release of jconv

[Erik de Castro Lopo]

Paul Davis wrote:

> actually, 18 has been out for a few months, and now 19 has been "let go" as
> well.

I love to keep 'em guessing :-).

Basically there's a number of ways to keep up with this. There's
a libsndfile announce mailing list,

http://www.mega-nerd.com/libsndfile/lists.html

I also announce on freshmeat and there's a libsndfile RSS feed on
my blog:

http://www.mega-nerd.com/erikd/Blog/CodeHacking/libsndfile/index.html
http://www.mega-nerd.com/erikd/Blog/CodeHacking/libsndfile/index.rss20
http://www.mega-nerd.com/erikd/Blog/CodeHacking/libsndfile/index.atom

Other than mailing individual developers, I don't see what else I can
do. Obvious Paul mainages to keep up :-)

Cheers,
Erik

PS : I do hope people realise that was tounge firmly in cheek.
-- 
-
Erik de Castro Lopo
-
Pastafarianism : http://www.venganza.org/
The intelligent alternative to 'Intelligent Design'.
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] New release of jconv

[Fons Adriaensen]

On Thu, Mar 12, 2009 at 11:03:54AM +1100, Erik de Castro Lopo wrote:

> Other than mailing individual developers, I don't see what else I can
> do. Obvious Paul mainages to keep up :-)

Since this is such a basic (meaning essential) lib for 
all things audio, maybe announce new releases on LAD ?
(Or did I miss it ?)

Ciao,

-- 
FA

Laboratorio di Acustica ed Elettroacustica
Parma, Italia

Be quiet, Master Land; and you, Professor,
will you be so good as to listen to me ?
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev

Re: [LAD] New release of jconv

[Erik de Castro Lopo]

Fons Adriaensen wrote:

> Since this is such a basic (meaning essential) lib for 
> all things audio, maybe announce new releases on LAD ?

Well I used to that ages ago and stopped because I was releasing
quite often.

Maybe I should start again.

Erik
-- 
-
Erik de Castro Lopo
-
J. Headley: "God, root, what is difference ?"
G. Haverland: "God can change the byte order on the CPU, root can't."
___
Linux-audio-dev mailing list
Linux-audio-dev@lists.linuxaudio.org
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev