IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
Binaural Sound For The Whole System or Internet Browsers
Wyld Stallyn
post Mar 29 2013, 15:50
Post #1





Group: Members
Posts: 13
Joined: 29-March 13
From: Fuldatal
Member No.: 107444



Hello, recently I found myself wondering if there is a way to apply a binaural DSP to the whole system, i.e. have the soundcard driver directly communicate with it to alter the stream from the get-go.

The reason I wonder about this is because I have very diverse music interests and tend to browse the web on YT etc. looking for more interesting music because that's where the choice of complete titles is greatest, and I find the idea of having to download everything where I find the stereo image lacking a bit cumbersome.

For more technical details, I use Opera on a Win 7 64-bit OS and currently own an ASUS soundcard with ASUS drivers, which I may or may not replace with an ODAC in the long run. Of course this would complicate this whole "let soundcard driver interact with plugin" thing a bit.

Perhaps there is also a dedicated program for this that can run indepentend of the soundcard driver? And no, the Dolby Headphone settings are not an option as they are incredibly annoying.

This post has been edited by Wyld Stallyn: Mar 29 2013, 15:56
Go to the top of the page
+Quote Post
slks
post Mar 30 2013, 09:40
Post #2





Group: Members
Posts: 408
Joined: 31-March 06
From: Houston, Texas
Member No.: 29046



"Binaural" refers more to a recording process than anything else. Typically it involves recording the sound using a replica of a human head with microphones placed in the same position and orientation as human ears. Unless the recording was actually made that way, you're stuck with trying to get the same effect with a DSP, and the results are going to be similar to what you'd get with the Dolby processing. It's kind of like upmixing stereo to surround sound in that you're basically trying to re-create spatial separation which is missing from the actual audio data. Once you start doing that, you're into the world of guessing and compromises, and the results are never going to be perfect or even particularly good. (By my opinion anyway - some people seem to prefer the processed sound.)

Since you mentioned YouTube and stereo separation specifically, I've noticed a lot of tracks on YouTube are downmixed to mono. I'm not sure if YouTube was using a mono audio setting at one time, or if it's being done by the users. I've downloaded a few very rare songs from YouTube, simply because there's no where else to get them. But I know to expect, even before I rip it, that the sound quality won't be great.

Now for some history - in the 60s when stereo recordings were first becoming widespread, some lazy producers would turn a mono mix into a stereo mix by cutting the bass to the left channel and the treble to the right. The effect sounds like the fader knob in your car radio being misaligned, and the term used for these mixes was "fake stereo"... it was no substitute for a real stereo mix.

This post has been edited by slks: Mar 30 2013, 09:42


--------------------
http://www.last.fm/user/sls/
Go to the top of the page
+Quote Post
Wyld Stallyn
post Mar 30 2013, 18:35
Post #3





Group: Members
Posts: 13
Joined: 29-March 13
From: Fuldatal
Member No.: 107444



Yeah I know it's a recording process, but I just naturally said "binaural" instead of "appoximating binaural", just because when I think about more enjoyable stereo sound per headphones, binaural as-is is the big word in my mind.

The downmix is done on YTs side for lower-quality versions of the videos. Vidoes in "HD"-quality generally don't suffer from this unless the uploader or someone else down the chain has already mixed the original file thus.

I can't relate to what that sounds like because I don't even use a car. I'm not doubting that it sounds mediocre at best though. Either way, for some mixes a binaural approximation does provide a definite enrichment, even if just for the crossfeed eliminating that dreadful one-sidedness. It's not really a big deal since this occurs rarely for me (making my cumbersomeness-comment a bit silly in retrospect), but I've been wondering nonetheless.

And using the Meier-settings with the BS2B-plugin with Mediamonkey it's not actually a very crass effect.

This post has been edited by Wyld Stallyn: Mar 30 2013, 18:38
Go to the top of the page
+Quote Post
Mach-X
post Mar 31 2013, 06:39
Post #4





Group: Members
Posts: 273
Joined: 29-July 12
From: Windsor, On, Ca
Member No.: 101859



I've toyed around with such toys, even bought a USB sound blaster recon3d with thx studio pro, messed with rock box and its cross feed effect, foobar with it's binaural plugin, and every time with music they all produce the same unnatural stereo image collapse. The only one I found somewhat effective was the Dolby headphone included with my netbook. Even then I found it unnatural though it was the only one really effective at externalizing sound. Was great for movies though. Your best bet really is to turn such effects on when listening to early hard panned recordings then turn it off again cause they all make regular recordings sound like crap.
Go to the top of the page
+Quote Post
euphonic
post Mar 31 2013, 07:10
Post #5





Group: Members
Posts: 86
Joined: 12-February 06
Member No.: 27711



The obvious solution if you're listening on headphones is to select "Headphone Virtualisation" under the driver properties, assuming you're not on XP. On speakers, as far as I know, you'd be out of luck as far as a system-wide solution goes. For which YouTube resolutions correspond to which audio/video bitrates, see http://en.wikipedia.org/wiki/Youtube#Quality_and_codecs .

This post has been edited by euphonic: Mar 31 2013, 08:08
Go to the top of the page
+Quote Post
Wyld Stallyn
post Mar 31 2013, 12:55
Post #6





Group: Members
Posts: 13
Joined: 29-March 13
From: Fuldatal
Member No.: 107444



QUOTE (euphonic @ Mar 31 2013, 08:10) *
The obvious solution if you're listening on headphones is to select "Headphone Virtualisation" under the driver properties, assuming you're not on XP. On speakers, as far as I know, you'd be out of luck as far as a system-wide solution goes. For which YouTube resolutions correspond to which audio/video bitrates, see http://en.wikipedia.org/wiki/Youtube#Quality_and_codecs .

Well for my speakers I obviously don't need a binaural DSP. biggrin.gif Very interesting table though, thanks.
QUOTE (Mach-X @ Mar 31 2013, 07:39) *
I've toyed around with such toys, even bought a USB sound blaster recon3d with thx studio pro, messed with rock box and its cross feed effect, foobar with it's binaural plugin, and every time with music they all produce the same unnatural stereo image collapse. The only one I found somewhat effective was the Dolby headphone included with my netbook. Even then I found it unnatural though it was the only one really effective at externalizing sound. Was great for movies though. Your best bet really is to turn such effects on when listening to early hard panned recordings then turn it off again cause they all make regular recordings sound like crap.

I don't know really. I noticed signification coloration with the "default" BS2B-settings, but very little with the "Meier"-settings, which is why I use it with Mediamonkey. In fact I'd say I'd be hard pressed to notice anything without careful listening, which of course implies that there's probably some expectation bias involved. I agree that Dolby Headphone is suboptimal, which is why I stopped using it after the first few hours. I guess I shall have to live with occasional one-sided sounds at times I can't use my monitors (I forgot to mention those over the weekend because I've been in a different household, apart from them - out of sight, out of mind.).
Go to the top of the page
+Quote Post
AstralStorm
post Mar 31 2013, 13:59
Post #7





Group: Members
Posts: 745
Joined: 22-April 03
From: /dev/null
Member No.: 6130



Hey,

I've done that thing a long time ago, you will need:
1) VSTHost or another VST host capable of routing audio from and to an interface. In terms of payware, Xlutop Chainer works with ASIO4ALL, as does LiveProfessor.
VSTHost is the least stable one - may crash when adjusting the effects, while LiveProfessor tends to have UI issues. Chainer has a silly tiny non-scaleable UI on the other hand.
2) Virtual Audio Cable or a similar loopback driver. (I have only tried VAC) There is no free driver of such sort, but it's not too expensive.
3a) ASIO4ALL to "merge" the two VAC cards in a one ASIO driver. Maybe some other ASIO driver can work too, haven't tried. VSThost also works in DirectSound and MME mode, but it's noticeably high latency then in my system, minimal 630 samples.
3b) Alternatively, you can use a Windows build of Jack Audio Connection Kit with JackRouter ASIO driver. Works extremely well too. I unfortunately was unable to set the 1.9.9.5 build into multiple device mode, so had to use ASIO4ALL still. It may be slightly more stable. in such a configuration.
If you manage to get it running in multiple cards mode, let me know.

Now, you need 2 VAC cards (cables), then insert VSTHost between them with correct clock adjustment for your sound output, and engage an Audio Repeater (KS preferably) from the second one (output) to your card of choice. Both of the virtual cards need
I've been using that with VST version of bs2b with very good results.
So enable ASIO4ALL and pick both VAC cables. Set the input cable as the default system sound device.

The tricky part is finding the clock adjustment necessary for your card.
For example, my Audiotrak Prodigy Cube runs at 100.0015% speed, while HDA runs at 99.99875%. Leckerton UHA6S mkII runs at 100.00009%. To find the clock, engage Audio Repeter (KS) with short buffer (e.g. 20 with 20 parts). Either it will underflow on the virtual card quick or on the real one. If it does on real one, increase the adjustment on the virtual cards (both!), if it does on the virtual one, decrease the adjustment. The longer it takes to fail, the nearer you are the target adjustment. I recommend the use of bisection or golden ratio section.

It's recommended to set VAC interrupts per second to every 1 ms - the lowest latency setting. Unless it drops out of course.


--------------------
ruxvilti'a
Go to the top of the page
+Quote Post
markanini
post Mar 31 2013, 14:10
Post #8





Group: Members
Posts: 550
Joined: 22-December 03
From: Malmö, Sweden
Member No.: 10615



I use Virtual Audio Stream and it allows me to apply any vst effects to system wide. You also simply use foo_record and set the input source in your soundcard mixer to stereo mix and use any of the crossfeed effects avaliable to foobar2000. This introduces noticable a delay so not great for video. Also if you havent tried tried Xnors crossfeed your missing out!

This post has been edited by markanini: Mar 31 2013, 14:10
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 1st October 2014 - 23:12