29 Nov
2010
29 Nov
'10
1:19 a.m.
Hi All
I think I know the answer to this, but I was wondering if it's
possible to mix two separate audio sources in software. I'd like to be
able to play announcement-type audio atop the currently playing main
track. An example might be when driving down the Autobahn and a
traffic report temporarily pre-empts whatever it is you are listening
to. In my case, though, I'd prefer to keep the main audio playing, but
at a very low volume level compared to the short announcement..
My setup is a roll-your-own embedded Linux distro with kernel 2.6.28.
I'm using Alsa SoC audio driving an I2S output on my Marvell processor
(which is now working just fine, BTW). The I2S signals feed an FM
transmitter chip which is broadcasting audio to a nearby FM receiver.
The FM chip only has the one input so no mixing possible within that
part. Anyhow, considering I only have a single I2S output, seems to me
that I'd need to do the mixing somewhere upstream of that serial port
within the Linux machine. But there is no special hardware for this
and I think the answer is "no, there is no way to mix two separate
audio tracks without a DSP and another device driver". But I thought
I'd ask just the same.
Regards,
Rory
29 Nov
29 Nov
1:39 a.m.
On Sun, Nov 28, 2010 at 04:19:40PM -0800, Rory Filer wrote:
I think I know the answer to this, but I was wondering
if it's
possible to mix two separate audio sources in software.
This could mean two things:
* mixing the outputs of two or more apps driving the same
sound card, or
* mixing audio signals inside an application.
The first is possible using ALSA's dmix plugin or Jack.
The second is what almost all audio apps will do at some point,
it's an almost trivial operation and only requires multiplying
and adding sample values.
On a modern PC you can do this with hundreds of signals without
problem, it certainly does not require a DSP chip.
Ciao,
--
FA
There are three of them, and Alleline.
30 Nov
30 Nov
7:18 a.m.
On Sun, Nov 28, 2010 at 4:39 PM,
<fons(a)kokkinizita.net> wrote:
This could mean two things:
* mixing the outputs of two or more apps driving the same
sound card, or
Yes, this is the case. I had already tried that and found that the second
app couldn't open the card because the first one already had it open.
The first is possible using ALSA's dmix plugin or Jack.
I will look into this along with the other suggestions in posts subsequent
to yours.
On a modern PC you can do this with hundreds of
signals without
problem, it certainly does not require a DSP chip.
Very reassuring and I am looking forward to understanding how this will
work.
Thank you for your response!
Rory
29 Nov
29 Nov
1:42 a.m.
Hello Rory!
Of course you should be able to do that in software. You have your two
seperate sources, which should be able to go into Ecasound or Ardour (if
you're graphical and don't mind JACK) or any other more or less full weight or
light weight multitrack harddisk recorder.
There you create two tracks, one for each source and then you can use the
volume faders. with jack that would be easy. If you want it light weight and
still rather comfortable, you can try Ecasound or Nama, which sits ontop of
Ecasound and has both GUI and CLI (commandline interface). In Nama:
# add a track for the normal programme
add music;stereo;source jack_client_1
# look at the created track
sh
[find effect ID of volume, for example F]
# add a midi controller for the volume from 0-100
add_controller F km 1 0 100 1 1
# add track for the announcements
add voice;stereo;source jack_client_2
# look at the created track
sh
[find volume effect ID, example H]
# add midi controller for volume, from 100-0
add_controller H km 1 100 0 1 1
Now just connect a MIDI controller to your PC and use the modwheel, or
whatever is controller number 1. If it's turned/pushed up, music only, if it's
completely down, voice only.
Aroudr will do that job just as well, and probably prettier. There are other
apps. I think even just mixer apps. I think most of them prefer JACK.
Kindly yours
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
8:39 p.m.
On Mon, 2010-11-29 at 01:42 +0100, Julien Claassen wrote:
[snip] There are other
apps. I think even just mixer apps. I think most of them prefer JACK.
Kindly yours
Julien
Patrick Shirkey's jackEQ was (and I guess it still is ;) a simple, but
very usable and stable mixer and it comes with a cross-fade option,
because it's designed as a DJ mixer, ideal to fade between traffic
reports and to what we usual are listening too. FWIW,it's more common to
do this kind of ducking automatically by a compressor, but any kind of
mixer.
http://jackeq.sourceforge.net/
Best,
Ralf
30 Nov
30 Nov
7:38 a.m.
On Sun, Nov 28, 2010 at 4:42 PM, Julien Claassen <julien(a)c-lab.de> wrote:
Of course you should be able to do that in software.
<Snip>
Hi Julien,
My hardware is barebones - no graphical interface, but a console with a command
line interface. I am also limited in the amount of Flash and RAM space
available so
I'll be picking the most lightweight option I can find.
On Mon, Nov 29, 2010 at 11:39 AM, Ralf Mardorf
<ralf.mardorf(a)alice-dsl.net> wrote:
On Mon, 2010-11-29 at 01:42 +0100, Julien Claassen
wrote:
> [snip] There are other
Hallo Ralf
Between your, Julien's and Fons' replies I should be able to figure
out a possible
approach given the limitations of my system. All three of you have
offered suggestions
for how to do audio mixing in software and that's some good news I
wasn't expecting:
dmix plugin,
Nama overtop of Ecasound
jackeq
Now it's a question of sorting through these and seeing which one(s)
will work on my
platform. I'll also need to figure out how separate applications can
open the same
audio device - but that might become clear when I understand how these
mixers fit
into the system.
Many thanks to you all for helping me out!
Best Regards,
Rory
2:52 p.m.
On Tue, Nov 30, 2010 at 1:37 AM, Rory Filer <rflyer(a)gmail.com> wrote:
dmix plugin,
Nama overtop of Ecasound
jackeq
Now it's a question of sorting through these and seeing which one(s)
will work on my
platform. I'll also need to figure out how separate applications can
open the same
audio device - but that might become clear when I understand how these
mixers fit
into the system.
I think you've walked away with slightly the wrong message here.
There are two approaches to the situation of multiple applications
using one soundcard.
One involves using the capabilities built into ALSA itself. This means
making each application open a dmix device. Rather than telling them
to open "default" or "hw:0" or "front", you would tell the
applications to open, most likely, dmix:0
The other involves using a user-space server. The only two servers
worth considering are PulseAudio or JACK and they are quite different.
PulseAudio can be used even though the applications using it believe
that they are just using a regular ALSA device. This is way that quite
a lot of distros set up PulseAudio these days. By contrast JACK
generally requires that applications are written to use JACK. This may
not require that the application itself is written to use JACK - the
support for JACK might come from some intermediate audio API such as
gstreamer. There are ways to route most things via JACK (even regular
ALSA-using apps) - details can be found in the JACK FAQ at
http://jackaudio.org/faq where you can also find a brief explanation
of the differences between PulseAudio and JACK.
6:06 p.m.
On Tue, 2010-11-30 at 08:52 -0500, Paul Davis wrote:
On Tue, Nov 30, 2010 at 1:37 AM, Rory Filer
<rflyer(a)gmail.com> wrote:
[snip]
I think you've walked away with slightly the wrong message here.
[snip]
The only two servers
worth considering are PulseAudio or JACK and they are quite different.
[snip]
Hi Rory :)
jackEQ is a mixer for JACK audio clients only, unfortunately some apps don't have JACK
support, OTOH Linux audio studio software should use JACK and AFAIK all common apps do
support JACK audio.
Using jackEQ with compatible audio apps is comparable to analog stand alone audio devices,
e.g. you can connect virtual cables of your hard disk recorder and of another application
to the jackEQ mixer, while jackEQ is connected by virtual cables to the JACK sound
server.
You only need to mix the audio sources by the mixer, as you would do when using analog
equipment.
Again ;) IMO for your purpose using a compressor with a sidechain feature is the best way
to duck music, while listening to the traffic news.
I'm not using sidechain and it might be that this could be an issue for Linux, but
there are some howtos that seem to solve this issue, if there should be an issue,
didn't watch them myself,
http://www.google.de/search?q=sidechain+compressor+linux&hl=de&prmd…
.
Cheers!
Ralf
1 Dec
1 Dec
4:21 a.m.
On Tue, Nov 30, 2010 at 5:52 AM, Paul Davis
<paul(a)linuxaudiosystems.com> wrote:
There are two approaches to the situation of multiple
applications
using one soundcard.
One involves using the capabilities built into ALSA itself. This means
making each application open a dmix device.
Bingo! I've been playing with this and got it working using "aplay" on
my embedded platform. I can play two tracks simultaneously now and
hear them both. In order to get it working I had to resolve a configuration
problem by adding "/etc/asound.conf" and I copied one from a posting:
pcm.playback {
type dmix
ipc_key 2048
ipc_key_add_uid true
ipc_gid 1
slave {
rate 44100
pcm hw:0
}
}
The "ipc_gid" parameter was giving me trouble until I did this. My aplay
command works when I use "-D playback" on both commands. Simple
for you guys maybe, but quite a thrill for me :)
I tried to run my squeezeslave client this way too, but it wants to go
straight and hog the /dev/dsp interface. So when I try running my aplay
while squeezeslave is up I get
"audio open error: Device or resource busy".
And Squeezeslave only supports numbers when specifying output devices.
Run it with the -L option and you see
Output Devices:
* 0: /dev/dsp
I'm inclined to try and alter squeezeslave to accept an argument like
aplay's -D. Failing that, ...
The other involves using a user-space server.
... I might give this a try
Thanks for the help!
Rory
3:05 p.m.
On Tue, Nov 30, 2010 at 10:21 PM, Rory Filer <rflyer(a)gmail.com> wrote:
I tried to run my squeezeslave client this way too,
but it wants to go
straight and hog the /dev/dsp interface. So when I try running my aplay
while squeezeslave is up I get
Squeezeslave stupidly uses the old OSS API, not ALSA. Redirectly this
to a different ALSA device is a bit different, because its a "legacy"
API and only supported to try stop old stuff from breaking.
It is possible to change what ALSA h/w device "/dev/dsp" corresponds
to, but I don't know of a way to make it refer to a dmix
pseudo-device.
Basically, Squeezeslave is stupid and old and needs fixing and/or not
using. And for what its worth, I run a SqueezeServer and 3
squeezeboxen, so I'm not exactly immune to the joys of squeeze-fu :)
--p
4889
days inactive
4891
days old
linux-audio-dev@lists.linuxaudio.org
9 comments
5 participants
participants (5)
-
fons＠kokkinizita.net -
Julien Claassen -
Paul Davis -
Ralf Mardorf -
Rory Filer