<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns="http://www.w3.org/TR/REC-html40"> <head> <meta http-equiv=Content-Type content="text/html; charset=us-ascii"> <meta name=Generator content="Microsoft Word 11 (filtered medium)"> <style>  </style> </head> <body lang=ZH-CN link=blue vlink=purple style='text-justify-trim:punctuation'> <div class=Section1 style='layout-grid:15.6pt'> Hi, I found below letter and met same problem with new BT module “module-bluetooth-device”.<o:p></o:p> Mplayer play stream as <o:p></o:p> AO: [alsa] 44100HZ 2ch s16le (2 bytes per sample)<o:p></o:p> And BT headset works fine in ALSA.<o:p></o:p> <o:p> </o:p> When I choose to PA by <o:p></o:p> Load-module module-bluetooth-device sink_name=bt address=xx.xx.xx.xx.xx.xx profile=a2dp<o:p></o:p> the stream played at least 2x the speed as the letter said. The module picks up configure correctly as<o:p></o:p>     monitor source: 2<o:p></o:p>     sample spec: s16le 2ch 44100Hz<o:p></o:p> channel map: front-left,front-right<o:p></o:p> <pre>so I want to know is it a blueZ bug or PA bug due to Lennart said it’s a BT bug <o:p></o:p></pre><pre>in reply to below letter.<o:p></o:p></pre> <o:p> </o:p> <o:p> </o:p> <pre><o:p> </o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>I suspect this is some type of resampler problem rather than an actual<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>connection problem.  The headset is a GN6210H (which is a phone headset,<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>so mono earpiece and mono mic).  The only other oddity is that apart<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>from having only one channel, its sample rate is 8000Hz.<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'><o:p> </o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>I have it all correctly plumbed into alsa using bluez and if I just use<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>alsa, I can send sound to it with mplayer using<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'><o:p> </o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>mplayer -ao alsa:device=headset<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'><o:p> </o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>Since mplayer sees the device directly, it sends the sound stream as<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'><o:p> </o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>AO: [alsa] 800Hz 1ch s16le (2 bytes per sample)<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'><o:p> </o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>And everything sounds fine.<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'><o:p> </o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>When I plug it into pulseaudio with<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'><o:p> </o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>pactl load-module module-alsa-sink device=headset<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'><o:p> </o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>It correctly picks up the sample rate and channels.  However, trying to<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>switch an application to it results in dramatic speed increases and the<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>output being chopped to pieces.  If I set up pulseaudio only to have<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>this device and nothing else, mplayer plays the sound file at at least<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>2x the usual speed.<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'><o:p> </o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>Since the output has this:<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'><o:p> </o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>AO: [pulse] 44100Hz 2ch s16le (2 bytes per sample)<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'><o:p> </o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>I suspect the problem is somewhere in the pulseaudio resampling down<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>from 2 channels at 44100Hz to a single channel at 8000Hz.  I've tried<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>forcing different resampling methods with slightly different sounding<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>results but still playing at at least 2x the speed.<o:p></o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'><o:p> </o:p></pre><pre style='margin-left:31.5pt;mso-para-margin-left:3.0gd'>James<o:p></o:p></pre><pre><o:p> </o:p></pre> <o:p> </o:p> Zhang Xin(Wing)<o:p></o:p> Intel, OTC ShangHai<o:p></o:p> <o:p> </o:p> </div> </body> </html>