speech-dispatcher grabs alsa hw device

Try blacklisting the hdmi by creating a file called blacklisthdmi.conf in /etc/modprobe.d/ with the following line in it:
blacklist snd_hda_codec_hdmi

Your use of /etc/asound.conf conflicts with PulseAudio's handling of
the sound devices. Please remove it, and choose the desired output
device from the Hardware tab of Sound Preferences after logging out
and back in.

On Wed, Jan 11, 2012 at 05:01, Dan Chen <email address hidden> wrote:
> Your use of /etc/asound.conf conflicts with PulseAudio's handling of
> the sound devices. Please remove it, and choose the desired output
> device from the Hardware tab of Sound Preferences after logging out
> and back in.

To clarify, I mean that you should remove /etc/asound.conf.

I can add the black list entry to blank out the HDMI card. I will report back on this.
This remedy, if it works, will not be optimum.. because a non-admin user will not be able to switch between sound sources.

I can remove the /etc/asound.conf file, not a problem. I was trying to get it to work. by creating it, and I proved that alsa was able to identify the desired sound card as the default. that's why I made it.

The second sound source was not appearing as a hardware choice in the applet. that's another reason why I created the /etc/asound.conf file. I'll remove it, log out, log back in. and I'm really sure the second hardware choice will not appear.
as it was before.. :-(

the "Sound Preferences" sound applet.. it only had one hardware choice. I'm calling it applet.. and I guess its real name is Sound Preferences.

ok. I had a few minutes this AM. I removed /etc/asound.conf and rebooted.
logged in as me (admin account) . the Sound Preferences hardware tab only shows the first device, as listed above.
which is the original reason for opening this bug report. :-)

I'll try black listing tonight and report the results.

What's the output from `aplay -l' ?

Could you run this command to attach all of the audio info (including the aplay)?
apport-collect 914575

the result of "aplay -l" is essentially the same as catting the /proc /asound/pcm listed above - but in a different format..

~$ aplay -l
**** List of PLAYBACK Hardware Devices ****
card 0: Generic [HD-Audio Generic], device 3: HDMI 0 [HDMI 0]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 1: Generic_1 [HD-Audio Generic], device 0: STAC92xx Analog [STAC92xx Analog]
  Subdevices: 0/1
  Subdevice #0: subdevice #0

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

apport information

I got some error about alsa-driver not installed.
I just did an apt-get update apt-get upgrade and reboot (new kernel)
tried apt-get install alsa-driver and "unable to locate package alsa-driver:..

what can I do next?

results from the black list test..

I added the blacklist entry for the hdmi device.
The results are: hdmi no longer appears in the Sound Preferences applet.
and unfortunately, nothing else is still not appearing... no hardware device is appearing now..

I'm gonna remove the blacklist file.

thanks!

alsa-driver is a source package. You don't need it unless you're doing something that requires rebuilding the sound modules.

At this point, it appears pulseaudio itself is the problem and not an underlying driver issue. so.. https://wiki.ubuntu.com/PulseAudio/Log

attaching a verbose pulseaudio file.. the log shows my session in Sound Preferences. clicking on hardware tab, disabling and enabling the hdmi card.. cant do anything with the other card, since its not appearing.. :0

attaching a verbose pulseaudio file.. the log shows my session in Sound Preferences. clicking on hardware tab, disabling and enabling the hdmi card.. cant do anything with the other card, since its not appearing.. :0

running pacmd from the command.. here's the output from list-sinks and list-sources

>>> list-sinks
1 sink(s) available.
  * index: 0
 name: <alsa_output.pci-0000_00_01.1.hdmi-stereo>
 driver: <module-alsa-card.c>
 flags: HARDWARE DECIBEL_VOLUME LATENCY DYNAMIC_LATENCY
 state: SUSPENDED
 suspend cause: IDLE
 priority: 9950
 volume: 0: 100% 1: 100%
         0: 0.00 dB 1: 0.00 dB
         balance 0.00
 base volume: 100%
              0.00 dB
 volume steps: 65537
 muted: no
 current latency: 0.00 ms
 max request: 0 KiB
 max rewind: 0 KiB
 monitor source: 0
 sample spec: s16le 2ch 44100Hz
 channel map: front-left,front-right
              Stereo
 used by: 0
 linked by: 0
 configured latency: 0.00 ms; range is 0.50 .. 371.52 ms
 card: 0 <alsa_card.pci-0000_00_01.1>
 module: 4
 properties:
  alsa.resolution_bits = "16"
  device.api = "alsa"
  device.class = "sound"
  alsa.class = "generic"
  alsa.subclass = "generic-mix"
  alsa.name = "HDMI 0"
  alsa.id = "HDMI 0"
  alsa.subdevice = "0"
  alsa.subdevice_name = "subdevice #0"
  alsa.device = "3"
  alsa.card = "0"
  alsa.card_name = "HD-Audio Generic"
  alsa.long_card_name = "HD-Audio Generic at 0xf0344000 irq 42"
  alsa.driver_name = "snd_hda_intel"
  device.bus_path = "pci-0000:00:01.1"
  sysfs.path = "/devices/pci0000:00/0000:00:01.1/sound/card0"
  device.bus = "pci"
  device.vendor.id = "1002"
  device.vendor.name = "ATI Technologies Inc"
  device.form_factor = "internal"
  device.string = "hdmi:0"
  device.buffering.buffer_size = "65536"
  device.buffering.fragment_size = "32768"
  device.access_mode = "mmap+timer"
  device.profile.name = "hdmi-stereo"
  device.profile.description = "Digital Stereo (HDMI)"
  device.description = "Internal Audio Digital Stereo (HDMI)"
  alsa.mixer_name = "ATI R6xx HDMI"
  alsa.components = "HDA:1002aa01,00aa0100,00100200"
  module-udev-detect.discovered = "1"
  device.icon_name = "audio-card-pci"
>>>

>>> list-sources
1 source(s) available.
  * index: 0
 name: <alsa_output.pci-0000_00_01.1.hdmi-stereo.monitor>
 driver: <module-alsa-card.c>
 flags: DECIBEL_VOLUME LATENCY DYNAMIC_LATENCY
 state: SUSPENDED
 suspend cause: IDLE
 priority: 1950
 volume: 0: 100% 1: 100%
         0: 0.00 dB 1: 0.00 dB
         balance 0.00
 base volume: 100%
              0.00 dB
 volume steps: 65537
 muted: no
 current latency: 0.00 ms
 max rewind: 0 KiB
 sample spec: s16le 2ch 44100Hz
 channel map: front-left,front-right
              Stereo
 used by: 0
 linked by: 0
 configured latency: 0.00 ms; range is 0.50 .. 371.52 ms
 monitor_of: 0
 card: 0 <alsa_card.pci-0000_00_01.1>
 module: 4
 properties:
  device.description = "Monitor of Internal Audio Digital Stereo (HDMI)"
  device.class = "monitor"
  alsa.card = "0"
  alsa.card_name = "HD-Audio Generic"
  alsa.long_card_name = "HD-Audio Generic at 0xf0344000 irq 42"
  alsa.driver_name = "snd_hda_intel"
  device.bus_path = "pci-0000:00:01.1"
  sysfs.path = "/devices/pci0000:00/0000:00:01.1/sound/card0"
  device.bus = "pci"
  device.vendor.id = "1002"
  device.vendor.name = "ATI Technologies Inc"
  device.form_factor = "internal"
  device.string = "0"
  module-udev-detect.discovered = "1"
  device.icon_name = "audio-card-pci"
>>>...

Your /dev/snd links are screwed up. I haven't seen that in my audio troubleshooting adventures, so I'll have to look into it more.

fyi -

root@neko:~# ls -laF /dev/snd
total 0
drwxr-xr-x 3 root root 240 Jan 12 19:50 ./
drwxr-xr-x 16 root root 4240 Jan 12 19:50 ../
drwxr-xr-x 2 root root 80 Jan 12 19:50 by-path/
crw-rw---T+ 1 root audio 116, 4 Jan 12 19:50 controlC0
crw-rw---T+ 1 root audio 116, 8 Jan 12 19:50 controlC1
crw-rw---T+ 1 root audio 116, 3 Jan 12 19:50 hwC0D0
crw-rw---T+ 1 root audio 116, 7 Jan 12 19:50 hwC1D0
crw-rw---T+ 1 root audio 116, 2 Jan 12 19:51 pcmC0D3p
crw-rw---T+ 1 root audio 116, 6 Jan 12 19:50 pcmC1D0c
crw-rw---T+ 1 root audio 116, 5 Jan 12 19:50 pcmC1D0p
crw-rw---T+ 1 root audio 116, 1 Jan 12 19:50 seq
crw-rw---T+ 1 root audio 116, 33 Jan 12 19:50 timer
root@neko:~# ls -laF /dev/snd/by-path/
total 0
drwxr-xr-x 2 root root 80 Jan 12 19:50 ./
drwxr-xr-x 3 root root 240 Jan 12 19:50 ../
lrwxrwxrwx 1 root root 12 Jan 12 19:50 pci-0000:00:01.1 -> ../controlC0
lrwxrwxrwx 1 root root 12 Jan 12 19:50 pci-0000:00:14.2 -> ../controlC1
root@neko:~#

From the pulseaudio log:

( 0.127| 0.000) D: [pulseaudio] module-udev-detect.c: /devices/pci0000:00/0000:00:14.2/sound/card1 is busy: yes
( 0.127| 0.000) I: [pulseaudio] module-udev-detect.c: Found 2 cards.

This means that something else is using the sound card, so PulseAudio can't access it. That's why it does not show up. If you don't know what that could be, the output of "sudo fuser -v /dev/snd/*" might help you figure out what it is.

@Dave Lentz: >Your /dev/snd links are screwed up

Thanks for helping out with triaging, but could you clarify this? They don't look screwed up to me.

I saw this and thought pulse was looking for the wrong /dev/snd
(alsa-lib)pcm_hw.c: open '/dev/snd/pcmC0D0p' failed (-2)
( 0.041| 0.000) I: [pulseaudio] alsa-util.c: Error opening PCM device hw:0: No such file or directory

@Dave, ok. That message is actually quite common for HDMI-only codecs, which are at device indices 3, 7, 8 and 9. So that would correspond to pcmC0D3p in the reporter's case (C0D3p = Card 0, Device 3, playback).
Have a nice weekend!

fyi -

# sudo fuser -v /dev/snd/*
                     USER PID ACCESS COMMAND
/dev/snd/controlC1: speech-dispatcher 1027 F.... pulseaudio
/dev/snd/pcmC1D0p: speech-dispatcher 1027 F...m pulseaudio

# ps -ef | grep 1027
106 1027 1 3 Jan12 ? 00:41:38 /usr/bin/pulseaudio --start --log-target=syslog
106 1333 1027 0 Jan12 ? 00:00:00 /usr/lib/pulseaudio/pulse/gconf-helper
root 12121 9971 0 18:31 pts/0 00:00:00 grep --color=auto 1027

I dont remember installing "speech-dispatcher"
this is a new PC and new install of ubuntu. Its precise because of wireless support..

papa@neko:~$ cat /etc/group | grep pulse
audio:x:29:pulse
pulse:x:115:papa,jasmine
pulse-access:x:116:papa,jasmine
pulse-rt:x:131:papa,jasmine
papa@neko:~$ cat /etc/passwd | grep speech
speech-dispatcher:x:106:29:Speech Dispatcher,,,:/var/run/speech-dispatcher:/bin/sh
papa@neko:~$ cat /etc/group | grep 29
audio:x:29:pulse
powerdev:x:129:
papa@neko:~$ cat /etc/group | grep pulse
audio:x:29:pulse
pulse:x:115:papa,jasmine
pulse-access:x:116:papa,jasmine

~# apt-get remove speech-dispatcher
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following packages were automatically installed and are no longer required:
  libtextcat-data libnux-1.0-common gir1.2-totem-1.0 gir1.2-totem-plparser-1.0 libtextcat0 libgtkspell3-0
  gir1.2-dee-1.0 libnux-1.0-0
Use 'apt-get autoremove' to remove them.
The following packages will be REMOVED:
  speech-dispatcher
0 upgraded, 0 newly installed, 1 to remove and 3 not upgraded.
After this operation, 1,155 kB disk space will be freed.
Do you want to continue [Y/n]? y
(Reading database ... 229191 files and directories currently installed.)
Removing speech-dispatcher ...
 * Stopping Speech Dispatcher speech-dispatcher [ OK ]
Processing triggers for ureadahead ...
Processing triggers for install-info ...
Processing triggers for man-db ...
Processing triggers for libc-bin ...
ldconfig deferred processing now taking place
root@neko:~#

I rebooted and ta-da! the analog sound card appears.

What is speech-dispatcher? is it part of precise?

@Luke Yelavich, if you read this, could you figure out why speech-dispatcher was hogging the sound card directly instead of going through PulseAudio?

Status changed to 'Confirmed' because the bug affects multiple users.

Speech-dispatcher is a text to speech system, mostly used for accessibility purposes at this time. You must have done something to cause speech-dispatcehr to start, because speech-dispatcher should only be loaded either if you enable it as a system service in /etc/default/speech-dispatcher, or you use a program like the Orca Screen reader, which causes speech-dispatcher to load automatically. In the case of speech-dispatcher running as a user, it should have used pulseaudio.

Could you please attach any logs you find in ~/.speech-dispatcher/log to this bug?

Thanks.

Actually I was not even aware that speech-dispatcher was installed on my
system, and I'm at a bit of a loss to understand how it came to be
installed. The problem only became manifest following upgrade to 12.04.
I also haven't used Orca. I have now uninstalled speech-dispatcher and
the "bug" has disappeared. I'm not sure whether any logs will remain
following removal of the package but I will look when I get home.

  DC

On 31/01/12 21:59, Luke Yelavich wrote:
> Speech-dispatcher is a text to speech system, mostly used for
> accessibility purposes at this time. You must have done something to
> cause speech-dispatcehr to start, because speech-dispatcher should only
> be loaded either if you enable it as a system service in /etc/default
> /speech-dispatcher, or you use a program like the Orca Screen reader,
> which causes speech-dispatcher to load automatically. In the case of
> speech-dispatcher running as a user, it should have used pulseaudio.
>
> Could you please attach any logs you find in ~/.speech-dispatcher/log to
> this bug?
>
> Thanks.
>

On Wed, Feb 01, 2012 at 08:21:17PM EST, David Clayton wrote:
> Actually I was not even aware that speech-dispatcher was installed on my
> system, and I'm at a bit of a loss to understand how it came to be
> installed.

it has been installed by default since as far back as at least Lucid I think, but since you didn't need ti and since it didn't cause you problems, it has been keeping out of your way.

if you can get logs, that would be much appreciated.

Luke

I don't know if this will be much help

There was no log, so I reinstalled speech-dispatcher. Initially sound
was OK, but when I started speech-dispatcher, speaker-test failed:

david@david-samsung:~$ speaker-test

speaker-test 1.0.24.2

Playback device is default
Stream parameters are 48000Hz, S16_LE, 1 channels
Using 16 octaves of pink noise
ALSA lib pcm_dmix.c:1018:(snd_pcm_dmix_open) unable to open slave
Playback open error: -16,Device or resource busy
d

log file is attached.

I then killed speech-dispatcher, but speaker-test still gives the same
error, and there is no sound on the speaker.

D

On 01/02/12 11:18, Luke Yelavich wrote:
> On Wed, Feb 01, 2012 at 08:21:17PM EST, David Clayton wrote:
>> Actually I was not even aware that speech-dispatcher was installed on my
>> system, and I'm at a bit of a loss to understand how it came to be
>> installed.
> it has been installed by default since as far back as at least Lucid I
> think, but since you didn't need ti and since it didn't cause you
> problems, it has been keeping out of your way.
>
> if you can get logs, that would be much appreciated.
>

On Tue, Jan 31, 2012 at 09:59:16PM -0000, Luke Yelavich wrote:
> You must have done something to cause speech-dispatcehr to start,...

All I did was:
- Installed 10.04 (netbook edition)
- Upgrade to 12.04

> ... because speech-dispatcher should only
> be loaded either if you enable it as a system service in /etc/default
> /speech-dispatcher,...

Yes, it is enabled in /etc/default/speech-dispatcher

> ...or you use a program like the Orca Screen reader,
> which causes speech-dispatcher to load automatically.

gnome-orca is installed.

> In the case of
> speech-dispatcher running as a user, it should have used pulseaudio.

I see these processes running:

hans@netbook:~$ ps -ef | grep speech | grep -v grep
106 1978 1 0 15:02 ? 00:00:00 /usr/lib/speech-dispatcher-modules/sd_espeak /etc/speech-dispatcher/modules//espeak.conf
106 2011 1 0 15:02 ? 00:00:00 /usr/lib/speech-dispatcher-modules/sd_dummy
106 2023 1 0 15:02 ? 00:00:00 /usr/bin/speech-dispatcher --pid-file /var/run/speech-dispatcher/speech-dispatcher.pid

User 106 is "speech-dispatcher".

> Could you please attach any logs you find in ~/.speech-dispatcher/log to
> this bug?

I attached
/var/run/speech-dispatcher/.speech-dispatcher/log/speech-dispatcher.log

Could you please attach /etc/speech-dispatcher/speechd.conf?

Thanks

Hans, for now, you can stop speech-dispatcher from running by editing /etc/default/speech-dispatcher, and set RUN to no.

I still think we have an upgrade problem from lucid to precise with speech-dispatcher here, I'll take a closer look.

On Wed, Feb 01, 2012 at 10:14:00PM -0000, Luke Yelavich wrote:
> Could you please attach /etc/speech-dispatcher/speechd.conf?

It is attached to this message.

> Hans, for now, you can stop speech-dispatcher from running by editing
> /etc/default/speech-dispatcher, and set RUN to no.

Good to know. However that is only a workaround.
I even had deinstalled the package "speech-dispatcher"
but it returned to me at the next dist-upgrade.

here is the one file - speechd.conf

I un-installed speech-dispatcher. there is no user directory ~/.speech-dispatcher for me or the other user on this laptop.

Ok thanks all. Nothing conclusive is shown in the conf and log files. I'll get a lucid install set up and attempt an upgrade to precise, and see what happens.

 affects ubuntu/speech-dispatcher
 assign themuso

This bug was fixed in the package speech-dispatcher - 0.7.1-6ubuntu3

---------------
speech-dispatcher (0.7.1-6ubuntu3) precise; urgency=low

  * debian/speech-dispatcher.postinst: Make sure speech-dispatcher is not
    enabled as a system service for users upgrading from lucid. The code
    in question has been dropped, and should have been done so a long time
    ago, at least for Ubuntu. (LP: #914575)
  * Tweak gbp.conf for precise
  * debian/rules: Do not re-run the configure script as a dependency of the
    clean target. It is not needed, and makes dealing with packaging in git
    a pain.
  * Do not copy fresh copies of config.{sub,guess}, we run autoreconf.
 -- Luke Yelavich <email address hidden> Mon, 06 Feb 2012 13:48:00 +1100