Day 2 Keynote Presentation

Hey everyone,
You can download the SDK samples with more code and scenarios for the topics covered in the talk at the links below:
AudioGraph: [Github Link]
MIDI: [Github Link]
MediaComposition: [Github Link]
If you have any questions or comments, please don't hesitate to reach out to Jason and me.
Hi Bala,
will the MIDI sample code work with the Launch Pad Midi controller? And will it work on the Raspberry Pi with Windows 10?
Thanks
@tellingmachine
It works with Launchpad S and Launchpad Mini. Yes, it works with Raspberry Pi with Windows 10.
If you want the actual demo Bala used, I'll post that. It's very close to this:
http://10rem.net/blog/2015/01/06/interfacing-the-novation-launchpad-s-with-a-windows-store-app
Pete
@tellingmachine: Like Pete said, the MIDI sample code will work with all USB MIDI controllers. It will also work with the Raspberry Pi 2 with Windows 10 (or any device that supports the Universal Windows Platform).
We will be releasing the code that was used for the actual demo in the talk on the GitHub link mentioned in the end. I'll post a link in the comments once it has been posted.
@pete, @bala,
Very cool stuff.
Thanks
Seems that we're getting closer to get (someday) into Windows a user level AudioGraph style rerouting between apps, hardware and virtual ports all as connectable nodes.
Been hoping for something like that for years, once low-latency and multi-client MIDI was done properly, which looks like we're finally getting with Windows 10! :)
Speaking of virtual ports, (audio) interface aggregation would be something useful too, btw, to group and present various interfaces as a unique multi-channel one to use with apps only capable of dealing with 1 interface.
Would be interesting to see all the above on... say Windows 10.1 ;) (i.e. or in my lifetime at least, eh)
@Koshdukai:Thanks for the useful feedback. We are always looking for our customers to tell us what you want to see in the platform.Those are interesting scenarios that we will certainly look into enabling! Keep the feedback coming!
AudioGraph sounds like old DirectSound concept! Remember IFilterGraph and Mapper, isn't it ? Anyway it seems you are still missing the audio aggregation and applications interconnection... to get inspiration on this, i invite you to take a serious look on Voicemeeter.
Here's a brief article and the source for the simple drum demo from the Day 2 keynote.
https://blogs.msdn.com/b/johnshews_blog/archive/2015/05/08/build-2015-and-audio-effects.aspx
Pete
Jason & Bala,
I don't know if the audience had been having a long day or something, but let me just say that while watching this recording from Tasmania Australia, I was cheering.
Some amazing stuff in your demo. Thank you to the Microsoft team for all the work.
Dan
@DownUnderDan:Thank you for the kind words, Dan! Please do not hesitate to reach out to us with feedback regarding the talk, scenarios or questions.
What are the chances of getting the samples to run on Windows Server 2012 in Azure?
@labs: The samples require Windows 10 to work since they use new functionality only available in Windows 10. Loading a Windows 10 Preview build in an Azure VM might work.
@bala
what are the chances of getting the samples to run on Windows 10 preview using „VS2013 Express for Windows“ (and not „VS2015 Community RC“). I ask this question because I temporarily had installed „VS2015 Community RC“ under Windows 8.1 and it crashed when calling create app package. Before (and now again) I was using (and am using again) „VS2013 Express for Windows“ for my Windows Store app GrafiCompo and „VS2013 Express for Windows Desktop“ for my other Windows desktop apps both parallel and it works fine. In my app I want to send MIDI commands using „Windows.Devices.Midi“
Gert
@Gert:Sorry about the late response Gert. I would not recommend using VS2013 to run these samples. The recommended and supported way to run the samples is only with VS2015. I recommend sticking to VS2015 on the Windows 10 preview release.
When will the white paper about low latency be released?
Keith
@Keith:We're still working on the whitepaper, finalizing details about various hardware configurations that could impact latency. Is there something specific I can help answer in the meanwhile? Please feel free to email me at balasiv_at_microsoft_dot_com
Hi
Will it be possible to have the audio layer of a video be an input node of an audiograph?
In short: can you play a video and do things with its audio track?
Thanks
@Flow: Great question. Yes - you can set the FileInputNode to take a video file as input and AudioGraph will handle the audio track of the file. All container/codec formats supported by MediaFoundation can be used as input.
If you encounter issues, do not hesitate to let me know.
Hi all,
I wanted to let everyone know that the whitepaper on audio latency in Windows 10 is now live here: [MSDN page].
The whitepaper contains more details and goes into greater depth into the topic than the talk at //build 2015. If you have any questions, comments or suggestions, please do not hesitate to reach out to me. I can be reached at balasiv_at_microsoft_dot_com