maxfaction
New Member
Same.
Question / Help No Audio from Media Source
- Thread starter Swizzle
- Start date
- Status
JanH
New Member
Still a feature I really like to have in. Just rearranged the audio setup completly. Mixer, Audition, VAC and sorts. So all missing now is to be able to hear the sound of any media played in OBS Studio/MP.
Maybe set it up in a way we can use VAC to direct sound?
Still awesome work done Jim.
Maybe set it up in a way we can use VAC to direct sound?
Still awesome work done Jim.
Chris Mchugh
New Member
The inability to monitor audio before it goes is an essential, basic function of any broadcasting software or hardware. The lack of this feature renders OBS useless.
A Colder Vision
New Member
Respectfully, Osiris, it may be nonsense to you, but to some of us, it is very important. Of course tons of people use it without that feature. Many are new adopters that never had it in the first place to lose it. Don't let that become a complacency issue, or a reason to ignore the issue all together. As many are used to not having the feature to begin with, there are many of us that were used to having it, that now don't. Just not everybody comes to the forums to complain about it. In fact, I probably wouldn't have commented on it if it hadn't been for your seemingly dismissive quip. I'm not meaning to come off as an asshole, but not having this feature has been quite bothersome and quite frankly, a hassle.
As a Twitch streamer, I've created a custom raid responses that involves fireworks synchronized to music. But since I am unable to hear the audio cues in the video that I need to hear to say and/or do certain things I need to do at certain times, I am forced to combine and operate between iTunes and the media source in OBS to achieve the desired effect, which leads to inconsistent results trying to sync the two. I often don't use the raid response anymore as it has become too cumbersome to launch iTunes, launch the video, then go back and shut down iTunes so it's not chewing up unnecessary resources. Viewers don't want to see a streamer fiddling with OBS other than the time it takes to switch scenes. Anything longer than that is unacceptable.
Over a year ago, it was stated that the "fix" was two patches away, yet all we keep getting is "it's planned", or some variant thereof. This feature was in OBS before it went MP/Studio, so I don't see why it wasn't included when MP/Studio was released. Either way, It would be awesome to get a response other than "it's planned." I love OBS, but this is one aspect which has been completely ignored.
As a Twitch streamer, I've created a custom raid responses that involves fireworks synchronized to music. But since I am unable to hear the audio cues in the video that I need to hear to say and/or do certain things I need to do at certain times, I am forced to combine and operate between iTunes and the media source in OBS to achieve the desired effect, which leads to inconsistent results trying to sync the two. I often don't use the raid response anymore as it has become too cumbersome to launch iTunes, launch the video, then go back and shut down iTunes so it's not chewing up unnecessary resources. Viewers don't want to see a streamer fiddling with OBS other than the time it takes to switch scenes. Anything longer than that is unacceptable.
Over a year ago, it was stated that the "fix" was two patches away, yet all we keep getting is "it's planned", or some variant thereof. This feature was in OBS before it went MP/Studio, so I don't see why it wasn't included when MP/Studio was released. Either way, It would be awesome to get a response other than "it's planned." I love OBS, but this is one aspect which has been completely ignored.
FragManSaul
New Member
This feature would be massively helpful for multiplatform streams. I use 4x OBS Studio programs to stream to different platforms and I have a dedicated Fullscreen projector for green screen overlay effects to those different screens. It would be nice for these green screen effects to contain audio by way of playing onto the desktop audio sources, It would also help a lot for me to hear them too. Thanks for all the hard work and I'm very excited for this feature coming out!
Attachments
I was merely responding to the claim that not having this feature renders OBS useless, since that is a personal thing. I know perfectly well that it would make it a whole lot easier for people dependent on audio cues.
And I can't give any other answers then "It's planned", because that's all I know about it, i'm not the dev.
Other things were considered higher priority then audio monitoring.
Hi guys.
If ur media source doesn't have an alpha channel (not tried with alpha, maybe can works...), You can do an HTML with that media file (using the video div and autoplay) and then add this to OBS like browser plugin (local file), I did this for some videos with audio and works :D
HTML (example):
You can add width, height and loop to the video div.
PD: Sorry for my english.
If ur media source doesn't have an alpha channel (not tried with alpha, maybe can works...), You can do an HTML with that media file (using the video div and autoplay) and then add this to OBS like browser plugin (local file), I did this for some videos with audio and works :D
HTML (example):
HTML:
<!DOCTYPE html>
<html>
<body>
<video autoplay>
<source src="UR_MEDIA_SOURCE.mp4">
</video>
</body>
</html>You can add width, height and loop to the video div.
HTML:
<video width="1920" height="1080" loop autoplay>PD: Sorry for my english.
Last edited:
John Hussey
New Member
Awesome work Jim. Also throwing in my vote for audio monitoring.
cmarcusdimarco
New Member
Just a quick word to all who come here, this worked extremely well for me. Can't quite get the loop yet, but as far as a hotfix goes, this did wonders. Huge thanks!
Do what you say you're going to do ^
Then hire more people or recruit more people who'll work for free. Having to FORCE yourself to micro-manage everything is counter productive and simply not fair for the developer, and the people the developer makes promises for.
We as streamers have an expectation to uphold when expected to use this software from individuals who want/need it to be used. I've worked in customer service for a long while now, and when something is stated to be done by a set time, then that means it's going to be done. Otherwise you're a liar and therefore cannot be relied on to fulfill anything of value for us in the future.
The people who use your software have a right to have their wishes and demands fulfilled in return of using your servies. If a siginificant portion of people wish to have this option coded into the program, then make it a prioriy over your "own agenda" of whats important. The developer needs to realize that his opinion of what should get worked on is just as important as everyone elses.
The demand has been made for well over a year, and all we get is stalled off over and over when the original statement was "two patches". If the OP preferred to take longer, then too bad. He gave us a time frame, now we expect him to uphold it. Especially when we as streamers need to be capable of hearing our audio to be able to put it into sync with other layers of audio tracks. People expect good quality entertainment and we need good efficient programs to use so we can have better quality output for our streams.
This post has proven to show that a siginificant portion of us all wants this option done. So please fulfill our wishes since we are using YOUR PROGRAM. If you continue to blow us off, it'll only lead to more and more people becoming fed up with the misleading and empty promises. There's my two cents.
Feel free to defend yourself in anyway necessary, it doesn't change the situation that what I say is factual and truth. I don't post this to rally up anyone or anything. I simply say what needs to be said, because this is ridiculous.
Last edited:
c3r1c3
Member
That's an excellent idea! How much money are you willing to put into having this feature added/developed? Considering that a good audio (C/C++) developer makes about $100+/hr you'll need to put out/raise a lot of money to be able to afford that person. (I estimate the work to be between 20 and 80 hours of work, but it can depend on the programmer and other criteria).
As to getting more free programmers, please let everyone you know who can code about OBS and encourage them join in! Also if you happen to know any excellent audio C/C++ programmers who will work for free, PLEASE forward their information to us.
Last edited:
Jack0r
The Helping Squad
That is simply unrealistic. How do you want to HIRE people for FREE? We have a few moderators, including myself that "work" for free for OBS, but you need to tell us on which tree those free programming guys grow.
You obviously have not worked in programming or for open source projects. Features are always delayed, pushed back, hell, games are even released as triple A titles for 70$ without features that were promised. Also planning for a certain time, but then not being able to fulfill it in that time frame is not a lie, it simply did not work out as planned.
Wrong. Neither Jim nor we need to fulfill any wishes. We do not even need to answer your forum posts, but we still do it. Which then delays work for OBS if Jim has to reply again and again. Of course we are happy if people use OBS and try to help them get it working whenever we can, but this does not change how the program is developed in general.
If for a certain features, a certain big change is needed this will take a big amount of time and might have to be pushed back.
Your expectations seem to be very wrong. You got an estimate of 2 patches which is not saying: its ready on the 1st day of august or you can shoot me. If you keep an eye on the github and check the releases, Jim is constantly working on fixing bugs, adding features and trying to stay sane.
Might want to think about switching to a paid software. You will have the same results, delayed features, "promises" that might not be fulfilled in time. But hey, then you at least pay for it and have a right to demand support. Never saw a software that allows you to demand feature requests, but hey, enlighten us.
We do not need to offer a forum, community chat, twitter, video guides or even talk to you. But we still do it.
Saying its planned is wrong, giving a possible time frame is wrong, saying nothing is wrong. What is he supposed to do? (Of course your answer will be: "add my feature I need")
Of course Jim would love to add any feature demanded by any amount of users as fast as possible, but its simply not like that in reality. He already acknowledged that the feature is needed and will add it as soon as he is able to work on the whole audio system.
No need to defend at all. But we still reply to such forum posts that eat our time and help no single person at all.
I totally understand that the feature is useful for certain situations and that people want it. But making these forum posts will get you nowhere near a result, it even scrapes of time that could be spend on adding the features you ask for. And let me make it clear again. If you think you can demand ANYTHING from us you are using the wrong software.
c3r1c3
Member
People do NOT have the right to have their wishes fulfilled. They can make a request from someone else (and the other party has no obligation to do it), or do it themselves. Also OBS does not provide any services. OBS is software only.
Either way it's something that is provided for free, so please don't think that you can demand anything. In the world of open source software (which OBS is) you can either code, pay someone to code, or ask nicely. You have done none of these in your post.
The developer wanted to add this feature before adding DX12 support, but ran into some roadblocks (explained further below), and the 'masses' wanted/needed Vulkan/DX12/UWP support much more then they needed audio output from the media source, so features are being added to OBS as per your own personal criteria. Why are you complaining about us following your stated ideals?
If you need such a feature there are several ways to get it, including paying an (external) developer to add it, buying hardware and/or software to allow for such, and/or looking into different workarounds. Also if OBS doesn't fulfill your needs on a fundamental level, then I recommend that you look into other software. That is what professionals do... spend the time and money on the software and hardware they need to get the job done right now.
If you want the highest quality (as you stated in your post), you'll need to use several programs along side OBS (many not related to putting out video), services (like TwitchAlerts and other such stuff), and lots of good & expensive hardware... right now.
Now to address some of the technical parts of this issue:
Basically Jim has been evaluating different pre-existing libraries that handle audio for a complete routing solution, but all of them have various issues (not multi-platform, buggy, poor performance, not GPL2-compatible, etc.) and has been looking into coding this feature from scratch... that means it could take a while. Progress is being made on this front, but 'just adding an extra output' isn't as simple as it seems on the surface (at least to non-programmers). So while the feature has been delayed, it is still a feature very much wanted by Jim.
So please remain calm, direct your energy into getting other developers on-board to help with this, consider helping out in the forums and live chat so that more devs have more time to code with OBS, and ask/consider how you can work around this missing feature until it is added.
Last edited:
I understand that you're frustrated. I realize you may need this feature for what you may be doing or trying to do, but you're really not looking at it from our perspective. You're letting your emotion influence what you say too heavily here and it's sort of blinding your perspective. Although the others have said "one man can only do so much" or "there are higher priorities", that's not exactly a clear/transparent explanation. I'll freely explain what's currently going on.
Audio monitoring is actually one of my highest priorities, but there are good reasons why it struggles to break through and become my "current" or "highest priority" item. Firstly its a sub-project that will require a month approximately, where as most other high priority items are much, much less.
First of all to the suggestion of hiring/finding someone to do it: we're not a company, it's a completely free project, so "just hire somebody" or "just find someone willing to do it" isn't how it works. I wish it were that easy. Secondly, even if I could hire someone, finding someone qualified enough for this particular item would be a task in and of itself. Qualified developers/engineers do not grow on trees, paid or not, especially for this particular feature. Also, I am definitely not micromanaging everything, in fact a lot of people are helping out with a lot of different things (and every single one of them are awesome for doing so). Things they are actually capable of helping with. I am grateful for every single awesome individual who helps out. Unfortunately this is not one of those things they can help with very easily.
Very few people are experienced enough for some of the work that needs to be done. This sort of feature requires a lot of skill and experience that few people have. It's one of many things that fall on my shoulders. Everyone is helping best they can, but when it comes to the more complex or lower level stuff, it typically falls on me. It's not just as simple as "output the audio" here, sources/scenes/transitions/mixing are all involved.
Originally for 0.17.0 I was going to implement audio monitoring because I thought I had finally found a nice, simple, compact audio output library I could possibly use to implement it quickly, and has support for all the target operating systems, but it turned out that it wasn't built to work with the MSVC (the compiler we use on windows). That compiler is vital for us and makes everyone's life less of a nightmare, so that option to speed up the development was out.
Then I have three other major features planned for 0.17.0 which I can reasonably do combined all in approximately the same time most likely:
- Capturing DX12 games (which I added in a recent hotfix early, still apparently isn't working at 100%, so I'll have to take a look at it), something I've promised I would get around to for a while due to the fact that many games are implementing DX12 support now. Took longer than expected but otherwise wasn't that difficult.
- Capturing windows store games. Microsoft is releasing a lot of xbox games on the windows 8/10 store, but capturing them isn't possible because they're sandboxed, so I had to come up with a method to get around the store's sandboxing. Mostly finished, but still requires a fair amount of testing and debugging before it's ready.
- Replay buffer (ability to continually record and then save the last [x] seconds or minutes of that recording). Another things I've been promising people and important for the project because it's a parity feature with classic. Not really all that difficult in the slightest fortunately.
So when it comes to things like this, I have these sort of choices on what I should be working on. The list of my priorities is not really that much of a queue -- I don't put things on my list and work on them in the order they're requested. I choose based upon my best ability what's best to work on at any given time. And I chose for 0.17.0 once again to delay it again because the costs just outweigh the benefits for this particular patch.
And this doesn't even factor in fixing bugs, don't even get me started on the windows 10 anniversary update issue recently that I spent weeks agonizing over. It was making the program just freeze up in the middle of streams/recordings with certain AAA games. I was agonizing it and investigating it until I found a workaround to the driver issue; I finally realized I could completely rebuild the entire graphics subsystem, and I spent another week making that workaround.. it was a lot of work. I had to put all objects in the D3D graphics subsystem in to a linked list, saved their internal data, then it goes through all graphics objects, and recreates/reinitializes every single D3D reference to restart the graphics subsystem and fix the issue. To the user they only see it as a small green hitch in the screen, but it prevented the program from crashing and stopping the streams/recordings. That took a while to fix/test/debug, but was critical to fix. It was truly was a serious pain, but it outranked all other priorities so I was forced to do it above all else ASAP because the program was crashing on people due to the driver bug with windows 10. Before that I was consulting nvidia, microsoft.. I actually managed to get in contact with them somehow. It was a nightmare situation, but it's finally over.
So you can see that development of this program (or any program) probably doesn't work the way you think it does. I have to use my time efficiently, and I have to do the best with what we have, and ultimately there have to be hard decisions that not everyone may agree with. Sometimes those decisions are done for my own sanity as well.
I will however try to make a compromise here: The libvlc has the ability to output to desktop somewhat easily. I'll see how much work it is to add that just as a temporary measure so people can have some way to output their videos with desktop audio. Although I'm not promising anything.
I just wanted to make things a bit more transparent because you seem frustrated. (And hopefully prevent anyone from responding to you angrily. And make the pain stop. Please make the pain stop. I need less pain right now.)
EDIT: ...And it appears three people responded before I had a change to finish writing this. This was sort of what I was trying to avoid. Let's all keep civil.
double edit: so this is how griffith felt.
Audio monitoring is actually one of my highest priorities, but there are good reasons why it struggles to break through and become my "current" or "highest priority" item. Firstly its a sub-project that will require a month approximately, where as most other high priority items are much, much less.
First of all to the suggestion of hiring/finding someone to do it: we're not a company, it's a completely free project, so "just hire somebody" or "just find someone willing to do it" isn't how it works. I wish it were that easy. Secondly, even if I could hire someone, finding someone qualified enough for this particular item would be a task in and of itself. Qualified developers/engineers do not grow on trees, paid or not, especially for this particular feature. Also, I am definitely not micromanaging everything, in fact a lot of people are helping out with a lot of different things (and every single one of them are awesome for doing so). Things they are actually capable of helping with. I am grateful for every single awesome individual who helps out. Unfortunately this is not one of those things they can help with very easily.
Very few people are experienced enough for some of the work that needs to be done. This sort of feature requires a lot of skill and experience that few people have. It's one of many things that fall on my shoulders. Everyone is helping best they can, but when it comes to the more complex or lower level stuff, it typically falls on me. It's not just as simple as "output the audio" here, sources/scenes/transitions/mixing are all involved.
Originally for 0.17.0 I was going to implement audio monitoring because I thought I had finally found a nice, simple, compact audio output library I could possibly use to implement it quickly, and has support for all the target operating systems, but it turned out that it wasn't built to work with the MSVC (the compiler we use on windows). That compiler is vital for us and makes everyone's life less of a nightmare, so that option to speed up the development was out.
Then I have three other major features planned for 0.17.0 which I can reasonably do combined all in approximately the same time most likely:
- Capturing DX12 games (which I added in a recent hotfix early, still apparently isn't working at 100%, so I'll have to take a look at it), something I've promised I would get around to for a while due to the fact that many games are implementing DX12 support now. Took longer than expected but otherwise wasn't that difficult.
- Capturing windows store games. Microsoft is releasing a lot of xbox games on the windows 8/10 store, but capturing them isn't possible because they're sandboxed, so I had to come up with a method to get around the store's sandboxing. Mostly finished, but still requires a fair amount of testing and debugging before it's ready.
- Replay buffer (ability to continually record and then save the last [x] seconds or minutes of that recording). Another things I've been promising people and important for the project because it's a parity feature with classic. Not really all that difficult in the slightest fortunately.
So when it comes to things like this, I have these sort of choices on what I should be working on. The list of my priorities is not really that much of a queue -- I don't put things on my list and work on them in the order they're requested. I choose based upon my best ability what's best to work on at any given time. And I chose for 0.17.0 once again to delay it again because the costs just outweigh the benefits for this particular patch.
And this doesn't even factor in fixing bugs, don't even get me started on the windows 10 anniversary update issue recently that I spent weeks agonizing over. It was making the program just freeze up in the middle of streams/recordings with certain AAA games. I was agonizing it and investigating it until I found a workaround to the driver issue; I finally realized I could completely rebuild the entire graphics subsystem, and I spent another week making that workaround.. it was a lot of work. I had to put all objects in the D3D graphics subsystem in to a linked list, saved their internal data, then it goes through all graphics objects, and recreates/reinitializes every single D3D reference to restart the graphics subsystem and fix the issue. To the user they only see it as a small green hitch in the screen, but it prevented the program from crashing and stopping the streams/recordings. That took a while to fix/test/debug, but was critical to fix. It was truly was a serious pain, but it outranked all other priorities so I was forced to do it above all else ASAP because the program was crashing on people due to the driver bug with windows 10. Before that I was consulting nvidia, microsoft.. I actually managed to get in contact with them somehow. It was a nightmare situation, but it's finally over.
So you can see that development of this program (or any program) probably doesn't work the way you think it does. I have to use my time efficiently, and I have to do the best with what we have, and ultimately there have to be hard decisions that not everyone may agree with. Sometimes those decisions are done for my own sanity as well.
I will however try to make a compromise here: The libvlc has the ability to output to desktop somewhat easily. I'll see how much work it is to add that just as a temporary measure so people can have some way to output their videos with desktop audio. Although I'm not promising anything.
I just wanted to make things a bit more transparent because you seem frustrated. (And hopefully prevent anyone from responding to you angrily. And make the pain stop. Please make the pain stop. I need less pain right now.)
EDIT: ...And it appears three people responded before I had a change to finish writing this. This was sort of what I was trying to avoid. Let's all keep civil.
double edit: so this is how griffith felt.
Last edited:
It's not my job to advertise that for this development team. So suggest that to them, not me.
I said hire OR recruit. If you can't comprehend two different options simply, then lash your anger at yourself, not at the man asking questions or proposing ideas.
And no desire to even attempt to fulfill wishes or demands? What is this a totalitarian based system? Because you pose off to be some sort of dictatorial type of individual letting small scale politcs enravage your ego. Are we your slaves that have to abide by your bidding? That's simply horse manure.
And if you don't enjoy responding to forum posts, them shut your mouth and scroll passed. No ones forcing you too.
Last edited:
If it is of no use or help at all, then cease all forum posts and do not allow anyone to comment. But since we ARE allowed to comment, then be prepared to face our appreciation, or wrath. Either way, you bring it all upon yourselves. So answer for it, not weep and snap at the hands who help feed you.
I'm glad to see so many jump out from the shadows to lash out at me when I finally disturb the status quo. Sure, you all enjoy thanks and appreciation, but when dung hits the fan and you are the ones to blame, how dare someone come to you and share the discomfort of the consensus. Answer for your own mistakes, the developer should've never said, "two patches". If you don't like it when someone finally calls you out on your crap, then don't dig your own graves. I've said all I will say, feel free to lash yourselves out again. I'm done dealing with this posse.
Stop your entitlement please. Because you are entitled to nothing when it comes to OBS.
I mean the software is free and open source, made by people in their free time, why do you think you can demand anything?
Hands that feed "us"? Which hands are those?
You have probably noticed obs is free, so im not sure what feeding is going on here?
The dev gave an estimation of "two patches", an estimation implies a certain degree of uncertainty. But as Jim pointed out it turned out to take a lot more time then expected to implement proper audio monitoring and he will try to make the VLC source output to desktop.
Again, this software is free, you are not entitled to make any demands.
Want a certain feature? Ask for it, but there are no guarantees, code it yourself, hire/recruit someone else to code it. Demanding it is not an available option.
I mean the software is free and open source, made by people in their free time, why do you think you can demand anything?
Hands that feed "us"? Which hands are those?
You have probably noticed obs is free, so im not sure what feeding is going on here?
The dev gave an estimation of "two patches", an estimation implies a certain degree of uncertainty. But as Jim pointed out it turned out to take a lot more time then expected to implement proper audio monitoring and he will try to make the VLC source output to desktop.
Again, this software is free, you are not entitled to make any demands.
Want a certain feature? Ask for it, but there are no guarantees, code it yourself, hire/recruit someone else to code it. Demanding it is not an available option.
Last edited:
Lapppy
Member
Watching this thread, I just want to let you know that I very highly respect you for posting your thoughts, especially ones that will get you ridiculed on these forums. I think it is very important that the views and opinions of these forums are challenged even though I agree with them.
Also I thought that what Jim said summarized very well as to what goes on behind the scenes with development of this program and was very respectful about it. Huge shoutouts to you Jim, because there are not enough people who are nice to each other on these forums.
So let's all calm down, be nice to each other, and be patient.
Don't worry, audio monitoring will come soon. I would love that feature too, and I would like it right now. However, posting on the forums isn't going to get it made faster, and I can't code so all I can do is wait patiently. I've also been wanting UWP capture for a while, but I don't go around nagging the dev(s) for it because I know it will come eventually, and some things have more priority over other things. That's just how development works. If I really needed it right now, I would use another program.
And hey, Jim said he might look into getting a very simple version of audio monitoring working! I think that we should be excited that the dev(s) are listening at least a little, so it's really not as bad as you might think.
- Status