-
Notifications
You must be signed in to change notification settings - Fork 86
AAC encoding error #34
Comments
Please build from https://github.com/EHfive/pulseaudio-modules-bt/tree/aac_fix_dev . If issue not fixed, show me the AAC error log |
Thanks for the quick response and attempt at a fix! Unfortunately the error hit again, but I see you've added additional debugging output which will hopefully help:
|
I'm not sure if this will tell you anything, but I tested a few more times and the error is identical every time:
|
The log shows nothing wrong in buffers. I pushed a commit to aac_fix_dev which make it initializes necessary AAC encoder modules only. |
It seems like I'm running the latest version of fdk-aac (2.0.0), but I'll try git master after giving your update to aac_fix_dev a shot. |
I tested your update and still got the error. I then updated to the master branch of fdk-aac and got the error again. Downgrading to v0.1.6 of fdk-aac however seems to have resolved the issue (it's been running successfully for an hour now). If newer versions of fdk-aac (2.0.0 and above so far) are going to break pulseaudio-modules-bt, would it be worth statically compiling v0.1.6 until things are resolved upstream so other people don't have this issue as well? Either way, thank you for your help! |
Actually, I'm OK with fdk-aac 2.0.0. |
Still OK with fdk-aac git latest (2.0.0.r100.g95858d7) pulseaudio v12.2 s16le 2ch 48000Hz |
Strange, maybe there's something unique about my setup-- I suppose if others have the same configuration quirks and run into this problem they can read this issue and see that they need to downgrade. Thanks again for your help and this project! |
Just noticed your other comment-- for comparison, I'm running: pulseaudio v12.2 s16le 2ch 48000Hz |
With fix applied still E: [bluetooth] a2dp_aac.c: AAC encoding error, 0x60; frame_count:0, in_bufSizes:4096, out_bufSizes 1332, to_read:4096, to_write:1332, encoded:0 |
I was having the same issue with libfdk-aac v2.0.0. Downgrading to v0.1.6 seems to have resolved the issue. pulseaudio v12.2 |
My headphones were disconnecting with the same error every few minutes too. I toyed with the settings and switching to CBR with pulseaudio-modules-bt-git r141.3ecf1d7-1 |
That's a good tip. Changed the /etc/pulse/default.pa accordingly and
encountered no problems so far. 😀👍
…On Thu, Aug 22, 2019 at 04:05:27AM -0700, Schlaefer wrote:
My headphones were disconnecting with same error every few minutes too. I toyed with the settings and switching to CBR with `aac_bitrate_mode=0` seems to avoid the issue.
pulseaudio-modules-bt-git r141.3ecf1d7-1
bluez 5.50
libfdk-aac 2.2.0
pulseaudio 12.2
--
You are receiving this because you commented.
Reply to this email directly or view it on GitHub:
#34 (comment)
--
Bitte denken Sie an die Umwelt, bevor Sie diese E-Mail ausdrucken.
|
Hmm, I'm not sure this issue should be closed. It's definitely constantly happening with In my case, this error seems to consistently trigger within 30 min of Spotify playtime:
pulseaudio-13.0 I have bluez-5.52 and fdk-aac-2.0.1 available in unstable, I guess I could try those out. |
Hello, I'm also affected by this issue
My system is Arch linux, pulseaudio 13.0-3, pulseaudio-modules-bt 1.3-6, bluez 5.54-1, libfdk-aac 2.0.1-1 |
First of all, thank you for this project! I didn't think I'd ever be able to use bluetooth headphones on Linux before finding this.
Everything seems to be working correctly (with pactl showing the AAC codec in use), but after a certain amount of time (sometimes minutes, sometimes seconds), the pulseaudio output device automatically switches back to the default (line out) and journalctl shows the following error:
I can then switch the output device back to the bluetooth headphones and it'll work again for a few more seconds or minutes until the error appears again. I tried manually switching the output away from the bluetooth headphones to see if the error was triggered in the other direction, but no error messages appear in that case. The bluetooth logs don't show any brief disconnects or anything either.
Let me know if there's any additional output or tests I can perform to help!
The text was updated successfully, but these errors were encountered: