<html> <head> <meta content="text/html; charset=utf-8" http-equiv="Content-Type"> </head> <body bgcolor="#FFFFFF" text="#000000"> <div class="moz-cite-prefix">Dear Georg, thanks for the answers, I really like your 2nd solution. But the problem is, add a another module-loopback in front speaker path, which means I need a another pair of alsa drivers (as alsa-sink.0 and alsa-source.0 in the pic), that what I have written to the alsa device, then I can read the data back from the input alsa device. It`s not a valid solution with our current hardware. also another problem is that: we support dynamic switch the sink of the pa_stream (pa_sink_input_move_to) we have 3 way to playback current stream: 1. front speaker path only pa_stream -> alsa-sink.1 -> front-speaker 2. rear speaker path only pa_stream -> alsa-sink.0 -> alsa-source.0 -> alsa-sink.2 (within a module-loopback ) 3. front and rear speaker combine pa_stream -> combine-sink -> alsa-sink.1 -> front-speaker -> alsa-sink.0 -> alsa-source.0 -> alsa-sink.2 (within a module-loopback ) I`m not sure may be I can add a fake a module-loopback, just memory copy but wait for a while to simluate the lantency ? Thanks anyway! BR, Lixin 在 2015年05月06日 14:05, Georg Chini 写道: </div> <blockquote cite="mid:5549AF20.3000401@chini.tk" type="cite"> <meta content="text/html; charset=utf-8" http-equiv="Content-Type"> <div class="moz-cite-prefix">On 06.05.2015 04:17, golden wrote: </div> <blockquote cite="mid:554979A7.6030304@gmail.com" type="cite"> <meta http-equiv="content-type" content="text/html; charset=utf-8"> <meta http-equiv="content-type" content="text/html; charset=utf-8"> <pre style="white-space: pre-wrap; color: rgb(0, 0, 0); font-size: medium; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;">Dear PulseAudio folks, We are using PulseAudio in our automotive project, the progress is pretty good so far. I need some help for solving the sync problems between two speakers. </pre> The pic below is the data flow of the two speakers: <img src="cid:part1.06060709.04000208@gmail.com" alt=""> without any syncing method, what we hear from the front-speaker is ahead of the rear-speaker (real-spear is slower for sure), our goal is that make the rear-speaker almost sync with (or catch up) the front speaker. </blockquote> Hi, with your manipulation of the audio and with module-loopback you introduce quite a lot of latency into the first path. Maybe you can simply set the latency of alsa-sink.1 larger than that of alsa-sink.0, not sure if this works. Otherwise I would suggest to make the paths look more or less equivalent, that would mean putting a module-loopback into the front speaker path. You could also try and use my (not yet reviewed) patches for module-loopback at <a moz-do-not-send="true" href="http://thread.gmane.org/gmane.comp.audio.pulseaudio.general/22956">http://thread.gmane.org/gmane.comp.audio.pulseaudio.general/22956</a> which allow a better control over the latency introduced by the module. If you use the patches and have a module-loopback in both paths you should be able to fully synchronize them by adjusting the latencies of the two modules. Regards Georg <fieldset class="mimeAttachmentHeader"></fieldset> <pre wrap="">_______________________________________________ pulseaudio-discuss mailing list <a class="moz-txt-link-abbreviated" href="mailto:pulseaudio-discuss@lists.freedesktop.org">pulseaudio-discuss@lists.freedesktop.org</a> <a class="moz-txt-link-freetext" href="http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss">http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss</a> </pre> </blockquote> </body> </html>