This guy doesn't make tutorials.. this guy educates. Thats why he is the best! He made me understand Davinci Resolve, and now he made me understand sound settings. And i've watched tutorials 3days straight from a huge number of other youtubers and none managed to explain something except the basics which everyone knows..
This sir actually teaches us about sound setting. He explains everything so we can do much better. I am in the stage of testing and retesting, and I have watched this video more than three times for his explanations. This sir doesn’t just make tutorials; this sir educates. That’s why he is the best, This video should be included with you show "revelator iO 24 software". You explain everything so well and really go in-depth with it. The high-pass filter EQ for the game audio is an idea I had never thought of before. After watching your tutorial a few times to really understand it, I began to fix my audio, and I’m really surprised by how much better it sounds! Like WOW! Thank you sir, for your efforts., EposVox!..🙂
Hands down the best tutorial for streamers…. Audio is most important and this video explains every part of audio and how to control it and it not control you. I use 7 audio sources not all at once of course and this has guided me in the right direction to mastering my stream audio… thanks Epos
This video should be included with every OBS install. You explain everything so well and really goes to the bottom with it. The high-pass filter EQ for the Game audio is an idea Ive never thought of before. And after having watched your tutorial a few times to really understand it, I began to fix my audio and Im really suprised just How Much better it sounds! Like WOW! Thank you EposVox!Im gonna send this to my streaming buddies, really hope they follow it to!
Watching this one year later after I noticed my Livestream YT VOD was quiet and I can't adjust it in post and you made me realize my audio settings were just OK and now they are PERFECT. Great presentation of this too. THANK YOU!
I switched from Streamlabs to OBS Studio last March 1 because of your videos regarding Streamlabs controversy and I'm glad that I made the switch. I'll be watching this video later. Thanks so much for the free tutorials! :)
I'm so happy to see EposVox helping the streaming world start to take audio more seriously!! So the pro audio world uses this as the proper level input as a rule. -18 on a dbfs meter is the same as 0 on an analog VU meter which means anything over -18 average is a bit too hot. Sure small peak jumps over -18 are ok but as a rule setting your input gain to -18 and then pushing the output up if it's needed is the right way to go. This is why the Green or "Good" level in something like OBS is set to around -18. Now there are other factors but honestly it can be more confusing than helpful such as LUFs or Peak versus RMS metering. So stick to a -18 dbfs input and your audio will be clean with plenty of headroom.
I've been constantly tweaking my settings, but I wasn't going about the process the best way. This video definitely showed me a better way, and I really appreciate you taking the time to make this.
Glad I found this video. Took me 8 hours to setup obs overlays and everything but couldn't quite get the audio right. This helped me out alot. Thank you
120% agree with the first gain/compressor settings. Great job man, love seeing more of this info out there. you also hit the EQ levels the same as I say: Mic -5 to -10, Discord: -15, Game -20 to -25 (loudest sound), music (-25 to -35). Generally disagree with compressing the game audio for stream, it makes the sound seem different to users that might play.
Ok, so I have just been getting started streaming and I was soooo frustrated following others re-hashed guides with a lot of misinformation! I'm a musician, so I knew the OBS guides I was watching (*cough* Alpha Gaming *cough*), didn't seem right. Having an analog mixer before an audio interface is what I am used to. Every guide I have seen has encouraged me to crank the gain and only reduce it if you are having issues with the noise floor or noise gate threshold. I set mine to 75% just like you said and then added digital gain and I have never sounded better. I haven't even touched EQ! Coming from the world of guitars, where having a generous amount of gain really helps carve out unique tone, to streaming has been a trip! I can't believe how much information you offer. You don't just teach the steps, you teach the concepts, and I cannot begin to thank you enough! No shade meant for Alpha gaming, but they lack the substance you have. I should have started here.
Biggest thing for me soundwise was adding a compressor on my system and chat sounddevices for my GoXLR with EAPO. I can't live without it anymore. Everytime I see people struggling with loud games etc. I tell them this. Lifesaver.
The OBS color markers always annoyed me because it causes a lot of people to think their mics are peaking when they go into the red. I wish that "true peak" was set by default, because it sets the red zone around -3 dB and is much more accurate. By the way, I love the way you showed to set initial gain. I already did most of this (like setting the compressor and then a limiter after it) as well as compressing other audio based on when I speak. But it's cool to see the whole thing. I do hope that this video gets popular because SO many people on Twitch have inconsistent audio and it hurts to watch. Edit: My only critique, personally, is where you set your levels for background and game music, especially using the original OBS marks. Back in September of 2021, someone published "Recommendations for Loudness of Internet Audio Streaming and On-Demand Distribution". There's a few things in there, but the main thing is that it recommends most audio to be normalized to about -18 LUFS. This document inspired me (despite lacking normalization on OBS) to try to at least get the audio closer, so I generally don't allow my minimum audio to really be lower than -25 or -30 dB. So in my case, I often play music around -20 to -30 dB depending on the song or what we're doing in stream, game audio often sits in the lower part of yellow (which would be about -12 to -20 but is often just normal volume and can reach -5 or so), and then my voice usually sits around -5 to -12 dB. On top of that, I run ducked compressors on my music and game audio, only slightly though. This allows my voice to be easily heard, but for people to still get most or all of the original sound as well. What this also does is, mainly, it helps people that watch on TV or Phone. Often times, with most streams, the audio is so uncompressed that the music level compared to headphones is so low, that when someone talks or gets excited, my speakers are getting blown out. As far as how this translates to headphones, I actually think it helps a lot, especially for people that use it as background noise. The audio is much more "normalized" kind of (mainly consistent), so they can set a volume and not worry about peaks or it getting too quiet. Though that's obviously my own opinion, and the goal is that they are still separated, but much closer together. Edit 2: Ah, I guess you basically do that at the end. My bad, lol. Though, I do think it would've been nice to just suggest that everyone basically does ducking, and offer non-ducking as an alternate solution where you probably mention "true peak" and the levels there. Edit 3: One tip for anyone using a GoXLR, I prefer pulling in the music directly from the input, and not from any outputs. This bypasses the GoXLR and allows you to set your music volume independently. This gives you the benefit of ALWAYS having your music at the same level no matter what (I also have a streamdeck button to set it to 80% in Spotify), and you can then compress it slightly for normalization. It's very handy.
This is easily one of the best videos around when setting up both mic and gaming audio. I had issues with my game audio either being too loud with my voice or my voice booming over the game audio and finding out Ducking is awesome. Thanks, awesome video
THANK YOU. i remember wanting to find a video like this months ago, however many videos didnt exactly apply to a stream/gaming environment. you really hit the nail with the info i was looking for back then.
Hi all! I’ll skip right to it, wow does this work! Amazing settings. The only problem, and it’s not with this guide but with games, is that games have different base volumes. My mic maxes around 15-20mb which sounds good. But shooters are particularly loud. The dampening while speaking is a godsend. Without speaking, your game volume should max at 28-30mb. This was around -30 on the slider for bf4. Jedi survivor would require around -22 to reach the 30mb cap. Shooters are just particular loud. Hope that helps!
Man! You explain things that nobody I heard explains. You really did an amazing job bro!!!! I wish I saw this a while back before buying equipment. Thanks for the info!
This is an excellent video. So many of the "stream gurus" of yesteryear silently moved over to shilling sponsored gear which makes it impossible to find objective opinions. You're teaching critical stuff like gainstaging which streamers never find online because they think channels like podcastage are for audiophiles only. Subscribed and look forward to more :)
I do not know if anyone else had this. But the closed captioning went haywire at the halfway point. The first section was great, I am pretty sure you actually took the time to edit the CC. It looks like the time codes just ran off. It was after an add, and I am watching on an Android phone. Great video. Keep up the great work
This is the best audio guide i've seen, great job. Now my $30 Fifine sound awesome in OBS but would this work with something like EqualizerAPO so my comms on discord, zoom and in game chat benefits from it? (just the mic filters)
Has a very new streamer, who has a very basic understanding of audio, but really do want the best for my viewers in the live setting, and for produced content, this is amazing. Thank you so much!
Dude this is a splendid video! As someone who has seen many videos on this topic, I can say that this was by far the most helpful, accurate and instructional of them all. Congrats and thanks!!
9:45 The different filters, gains and so on in all the voice chat software annoyed me. I stumbled upon "Virtual Audio Cable" (VAC) (free version for non-commerical use). In OBS you select the VAC as monitoring device. You can then send your audio to output AND monitoring. Your video conference software can then use the VAC as input device. Thus all your finely tuned OBS settings can be used within every voice chat with the same gain, noise suppression and so on.
Superb work. Thank you. I’m having an issue with monitoring the mic audio, getting a slight delay in my headphones. I’d love to see your analysis of how to fix those issues.
Fun fact: you can use Equalizer APO to add VSTs system-wide so you don't need filters in OBS, yet have the same functionality, while equally making it work in other programs. Downside is that you don't have a visual of how these VSTs are applied, so if you need that you can test out settings in OBS first (or record and check in Audacity after each tweak) to then apply it.
Really, really good video. I love the insight into your new processing chain and when the gains come in, as well as the step by step on auto-ducking. Will definitely be coming back to this one!
I love this! Having studied audio engineering it always confuses me how people have their gain so high when they can fix it all in post and have plenty of control I have always ran my sound through a DAW especially with the GoXLR as the ASIO works really well
@@PurpleSpiritFoxFire not entirely, I use Reaper to process my sound as I’ve spent hundreds on plugins so I may as well use them. Then I route it to the Sample channel on my GoXLR mini then on the routing table for chat mic I have it ticked on only sample. Then it just mixes with my broadcast mix on stream. I still have fader control on my mic and muting, even bleeping works still. All with next to no latency. I’m sure Addie could explain in further detail if the interest was there!
@@PurpleSpiritFoxFiredo you have a GoXLR? If so the process Is fairly simple, if not it is doable but you can run into audio issues using virtual audio cables etc
This is (I think) going to help me a lot. I sing to backing tracks through an analogue usb mixer into OBS and I tried to add filters to my mic, but when I played the backing track, it was all choppy and impossible to sing along to. Obviously because the filters had affected the whole signal from the mixer. Thanks for this, I'm off to give it a shot lol 😉👍
Very, very, very cool video. I've been doing YT vids as a hobby for years and had already learned just about all of these, but that tooks months of mistakes and learning the hard way. One thing i really look forward to about this video, is OMG MAYBE NOW WE'LL HAVE LESS BG MUSIC DROWNING OUT THE PERSON TALKING! gd that's a pet peeve of mine. You even mentioned mic technique at the end...though you put it after the outro...that's bad because i think people are gonna click off of the video thinking it was over and that the end would only have "subscribe to me" crap at the end.
@@EposVox i get that, but i was trying to say people probably click off the instant they see the outro graphic - thinking there's no more information otherthan typical YT'ber "hit like and subscribe" plugs.
I love all your videos. For videocalls (zoom, g2m) to avoid the low voice volume I'm using all the knowledge I've learnt here as I'm sending OBS audio via a Virtual Cable with Audio Monitor plug in. I don't know if that is the best way but It's working atm.
Equalizer APO seems to also be a great free open source software alternative to using Voicemeeter/Steelseries Sonar, thought I'd mention it just in case
Alright, time for me to go back and lose my mind tweaking audio settings again! (I kinda love it though, and this will certainly be helpful) Thank you!😂👍❤
I'm an audio engineer and producer, and I never let audio go into the red. Red always means STOP. Apparently it's fine for gamers to overdrive audio levels well into the red, and not worry about what it sounds like on the other end of the stream. Yes, audio mixes are subjective--get ten engineers and ask their opinion, you'll get twenty different opinions, but audio levels are pretty standard. I just never let audio get into the red.
I mean, objectively if you’re not going past 0dB, you’re not clipping or overdriving anything. While what you said is generally sound advice (ha ha) that completely ignores the context of who designed the red and what it was for. Peaks being under -6db will be way too quiet for normal viewers and result in a horrible experience when they crank audio up to hear it and then click on any other video or stream.
I was wondering why my mic audio in OBS was showing just shy of red, but in the final mix I felt like it was kinda quiet. Looks like I got some re-working to do with gain and compression filters!
If you use a GoXLR would you recommend using the GoXLR's broadcast mix for the stream's audio and ignoring the audio filters (aside from a limiter) in OBS, or would you prefer to to keep each audio track separate so you can apply more audio filters (such as ducking) for each track in OBS for the stream?
If you want to use a full mix off your interface (I'm on a WaveXLR, so I use my full mix out of Wave Link) you can still get a "ducking" effect using gain > compression > limiting. Set your relative levels of (say) music, game, mic in ascending order...in my case, music is at something like* -32dB, game at -24dB, and mic at -18dB, all in RMS (not peak!). Now, obviously your music will be far too quiet and your game won't be very loud either. That's fine, because we'll throw like 7-8dB of gain on the whole mix. Now your mic will be too loud. That's fine, do a 2:1 compressor with a slow attack (like 500ms) and a slow release (like 1000ms), so you don't get unnatural "pulsing." But that compressor will push down the "quieter" parts of the mix any time the "louder" parts come in. So if you're not talking but the game is going, the game audio (at 6dB above music) will push down the music. If the game gets quiet, the music will come up (as the compressor releases). And if you talk, everything gets pushed down as the compressor attacks again. The limiter stage will prevent the gain at the front end from pushing your mic (or game) into clipping...you'll potentially get some subtle limiting on your voice (if you're being loud) for the first 500ms as the compressor attacks, but it won't be remotely noticeable, and the end result will be a nice effect where you're always front and center, but as "louder" parts peel away "quieter" parts will fade back in (with the 1000ms compressor release). All without needing access to sidechaining or a bunch of different sources in OBS. You can also add like 2dB of gain between compressor and limiter, to boost overall loudness, if you find your total mix isn't quite loud enough. You can actually get all of this rolled into one plugin, too. I personally use TDR's Limiter 6 GE, but that does cost money...it ran me about $30 on sale. You can also just as easily pick up (from the same TDR website) the original VladG Limiter 6 plugin, which does all of the above, only functions you'll lose are saved presets (outside of OBS) and the loudness monitoring on the final output stage. That and active support, obviously, because it's an obsolete plugin. It does work though, I've used it. Free and paid versions, respectively: www.tokyodawn.net/vladg-limiter-n6/ www.tokyodawn.net/tdr-limiter6-ge/ These plugins are the main reason I hate OBS's lack of a "final mix" that you can add plugins to, because a plugin like this on your final mix lets you achieve some really nice effects like I described above, and is great for managing your overall loudness levels and dynamic range, but kinda requires you to have everything all muxed into one final stream to apply it to. But for those with devices like GoXLR, WaveXLR, etc. this is a very nice option for mixing everything outside OBS and just applying one last filter to bring it all together. * - These measurements are actually in LUFS, not dB...LUFS are very similar to RMS dB, but not quite the same.
I personally have my chat as broadcast (because you can't use chat mic) and then system on chat mic, then music is pulled directly from the GoXLR input rather than any outputs. This allows me to set music in OBS completely separately, as well as apply chat and system sounds with different filters.
@Well Well Well >no matter what you do over time ... audio will desync Er no it won't. Maybe some of your audio or video is in 59.94 and some is in 60 FPS?
I use EqualizerAPO to apply all of my mic filters incl. compressor, limiter, and sometimes RNNoise(using vst thing) because laptop fan, to the microphone device directly, and all apps get the processed audio, without using virtual audio cable or another sound device or whatever. It's kinda spotty with realtek drivers though, I had to use generic microsoft drivers to make it work, but it works well when it does.
What should level should the windows microphone be at? I feel like this is common knowledge but no one ever addresses it and I want to know for my own sanity. Great video as always.
thanks for the video, there is a tiny mistake , when you are talking about the limiter, you say that its good to have the release at 60 db instead of 60 ms. xD
I've been having the problem in games like Apex, or which involve gunfire at all, with that sound peaking really badly and being too boomy. Do you have a suggestion for what to set up exactly to keep the game audio evened out? Your Apex clip sounded pretty good
Needed 60dB of gain on Presonus IO24 for Dynacaster to hit average peaks between negative -18 and -12dB. So much for turning my gain down (it's maxed out). Great video though. P.S - The quality of your videos has really gone through the roof as of late!
I think I had a fundamental misunderstanding of filter/signal chains. So... if your input after hardware gain is above 0, that audio is pretty much ruined and unrecoverable. Okay. But if I'm understanding this right, you can use a gain filter that would bring the audio above 0, and as long as you get it back below zero with compression and/or a limiter... it won't be destroyed? It's not destroyed before it reaches that point in the filter chain? If I'm understanding this right, this will merit completely reworking my filters. I've always put my additional "boost back" gain LAST in my chain, after compression/limiter, because I didn't want to clip early in the chain... but it sounds like I don't need to do that, and should put it first?
I've been using an OBS plugin by bozbez called "Win-Audio-Capture" and it's allowed me to set each individual exe file with a seperate audio capture without using Voicemeeter. That way I can tune everything individually through OBS. So before my friends in Discord sounded quiet to me but for Stream they were really loud. Now I just added an audio capture source targeting discord exe and I've turned them down on OBS.
Great tutorial, Thanks. I can see that you put the order Boost Back -> Compressor -> Limiter. If I have to add EQ, where should it be put in the sequence?
Do you have any recommendation for gaming headset with microphone? I have gaming headset Corsair HS80 ... my sound always on streams was fine but I notice that time to time, when I laugh loud or say something very loud it was going in that red line. Now last days it was getting even worse. I'm already using RTX voice for suppress the sounds around (background sounds) and in OBS it was now going in that red line. I literally followed up to your video and done everything the same, however where is BOOST BACK - instead of putting 16db ... I put -6.00db. and its sounds better than if I put 16db. Do you know why for me its sounds better with minus 6db instead of positive 16db? Because I put positive 16db it will be clipping and going into that red line. Do you know why?
Great video Epos, informative as ever! i'm struggling though in so many areas with my sound it's hard to know where to start first. Does this tutorial apply to condenser mics aswell? i have the yeti and i followed the section on bringing levels back up by adding gain filter set at levels to counteract peak levels lost (from compressor) as u suggested.Although i've added a limiter to avoid clipping, it sounds very blown out and really unpleasant on the ears. One question i always have is, do we set our mic inputs on windows to be 100 and control volume on a software/mic side (my mic has a physical gain ) or do we set input on windows relatively low and boost back up in obs with gain filters etc? or does it not matter which way you approach that? apologies for 20 questions, i'm never saitisfied with this damn mic ha! Great video though, very thorough as it has already been stated. You always come across very sincere and passionate about spreading the right information to the ppl that need it.
This applies to all mics. Right click that 100% slider in Windows and change it to dB If you’re using a XLR mic with an interface or a USB mic with its own gain dial, you want that at 0dB (which could be 100%, 50% or any other number sadly) If this is your only place to change gain, set gain here
@@EposVox Thanks for the prompt response. i've gone back to 50% on windows side as it's the only way i can kinda clean up the sound. i guess its abit of a balancing act huh. Thank you though, i plan to invest in a dynamic/xlr set up at some point in the future with an interface so hopefully i can yield better results there.
@@spenddollarzent ??? Yes? I'm only one human and only have so much time to reply to comments. I reply to all the ones I can/I think would be productive
I have access to the following equipment: DBX 286s (analog audio strip) Rode Procaster Pro (v1) Wave XLR Computer running Reaper (FX Chain) A good XLR mic (SM7B and RE20) I also have a Zoom F6 because I wanted the ability to record 32-bit float in the field, but it can double as an audio interface too (nice quiet preamps). I thought about connecting the DBX to this, but wasn't sure if that would negate the anti-clipping superpowers of the F6. I'm not trying to stream gaming or anything. Just content production in general. Is there a significant difference between analog filters like in the DBX or software based (the others). I believe I can have a copy of the audio recorded to SD card in the Procaster pre-fader (pre-fx?) as a backup, then apply fx manually in Reaper in post. I believe the Procaster also processes the fx on the device before sending to the computer to OBS or a recording app - this might be the ideal streaming setup in my case. Basically, I have a computer for production (recording & editing, not streaming) and a computer for streaming.
Hardware better than software. Less latency, which is important if you’re routing things to different places. Ideally you should consolidate to a high end interface that has DSP like that built in. If you’re more of an audiophile, you would then want to add a preamp even in front of the interface.
Followed this to the letter and my god you are a saint. I always wonder about my audio in streams and after testing all of this, I definitely have good headroom and better audio quality. Just to add, the ducking effect IS lowering the Broadcast Stream Mix in OBS, but my mic drops off to nothing or doesn't get recorded when I have that sidechain compressor enabled. Maybe I'm trying to duck something that shouldn't be ducked?
Keep in mind the lower your peak recording gain the more you are going to have to add gain up in post. This will make your room noise louder. You may start to hear your ac and untreated room more.
@@EposVox Maybe I expand my self poorly. It is true tho. Anytime you turn up a recording in post you are also turning up the room noise in that recording. If I have a recording where my voice peaks at say -12 db and I turn it up digitally in post it will also raise the room noise when im not speaking. Anytime you are adding gain you are raising your noise floor with your main signal.
Ok, but if you turn up gain on your interface instead, that same room noise comes up - plus the risk of added amp noise. Gain in post doesn't bring it up anymore than your amp gain would.
9:48 There is away to run OBS into other apps with a Virtuial Audio Cable, Voicemeter lets u download one for free and i set it up so i can run OBS into discord and VRChat so my mic doesn't sound dull or too quiet.
04:49 - 05:37 this is the most replayed part (for now) Then why should we set preamps to 75% and not 50%? I set my gain to the level where my noise floor is around -60dB (less than 3 o'clock) And it never hits -6dB 01:37 because I talk very quietly. Most of the time it would be around -27dB 02:17 - 02:24 It would be around 100% gain for me.
When watching this I think you're confusing Dynamic with Condenser, you say your gain dial will be lower if using a dynamic mic which isn't true, condensers have way more sensitivity so require less gain where dynamic generally needs more since they're mics that can take a beating, like on stage for example. I know this video is old but that point really confused me. To reach the points you want without using any OBS my dial is already at 75%
This guy doesn't make tutorials.. this guy educates. Thats why he is the best! He made me understand Davinci Resolve, and now he made me understand sound settings. And i've watched tutorials 3days straight from a huge number of other youtubers and none managed to explain something except the basics which everyone knows..
This sir actually teaches us about sound setting. He explains everything so we can do much better. I am in the stage of testing and retesting, and I have watched this video more than three times for his explanations.
This sir doesn’t just make tutorials; this sir educates. That’s why he is the best,
This video should be included with you show "revelator iO 24 software". You explain everything so well and really go in-depth with it. The high-pass filter EQ for the game audio is an idea I had never thought of before. After watching your tutorial a few times to really understand it, I began to fix my audio, and I’m really surprised by how much better it sounds! Like WOW!
Thank you sir, for your efforts., EposVox!..🙂
Thank you :)
Hands down the best tutorial for streamers…. Audio is most important and this video explains every part of audio and how to control it and it not control you. I use 7 audio sources not all at once of course and this has guided me in the right direction to mastering my stream audio… thanks Epos
Few minutes in and one already knows that this is it. Thanks a lot!
This video should be included with every OBS install. You explain everything so well and really goes to the bottom with it. The high-pass filter EQ for the Game audio is an idea Ive never thought of before. And after having watched your tutorial a few times to really understand it, I began to fix my audio and Im really suprised just How Much better it sounds! Like WOW! Thank you EposVox!Im gonna send this to my streaming buddies, really hope they follow it to!
The OBS Discord does link to his 5.5 hour masterclass.
Watching this one year later after I noticed my Livestream YT VOD was quiet and I can't adjust it in post and you made me realize my audio settings were just OK and now they are PERFECT. Great presentation of this too. THANK YOU!
this literally blew my mind when i realized i am not giving myself enough headroom for my audio AHHHHH
I switched from Streamlabs to OBS Studio last March 1 because of your videos regarding Streamlabs controversy and I'm glad that I made the switch. I'll be watching this video later. Thanks so much for the free tutorials! :)
I'm so happy to see EposVox helping the streaming world start to take audio more seriously!! So the pro audio world uses this as the proper level input as a rule. -18 on a dbfs meter is the same as 0 on an analog VU meter which means anything over -18 average is a bit too hot. Sure small peak jumps over -18 are ok but as a rule setting your input gain to -18 and then pushing the output up if it's needed is the right way to go. This is why the Green or "Good" level in something like OBS is set to around -18. Now there are other factors but honestly it can be more confusing than helpful such as LUFs or Peak versus RMS metering. So stick to a -18 dbfs input and your audio will be clean with plenty of headroom.
I've been constantly tweaking my settings, but I wasn't going about the process the best way. This video definitely showed me a better way, and I really appreciate you taking the time to make this.
Same I understood the concepts of the filters but his method gave me clean audio
This video was super helpful!
Glad I found this video. Took me 8 hours to setup obs overlays and everything but couldn't quite get the audio right. This helped me out alot. Thank you
120% agree with the first gain/compressor settings. Great job man, love seeing more of this info out there. you also hit the EQ levels the same as I say: Mic -5 to -10, Discord: -15, Game -20 to -25 (loudest sound), music (-25 to -35). Generally disagree with compressing the game audio for stream, it makes the sound seem different to users that might play.
-db were Surround sound volume on an Equalizer and a Receiver and a Studio monitor Speakers and Subwoofers.
Ok, so I have just been getting started streaming and I was soooo frustrated following others re-hashed guides with a lot of misinformation! I'm a musician, so I knew the OBS guides I was watching (*cough* Alpha Gaming *cough*), didn't seem right. Having an analog mixer before an audio interface is what I am used to. Every guide I have seen has encouraged me to crank the gain and only reduce it if you are having issues with the noise floor or noise gate threshold. I set mine to 75% just like you said and then added digital gain and I have never sounded better. I haven't even touched EQ! Coming from the world of guitars, where having a generous amount of gain really helps carve out unique tone, to streaming has been a trip! I can't believe how much information you offer. You don't just teach the steps, you teach the concepts, and I cannot begin to thank you enough!
No shade meant for Alpha gaming, but they lack the substance you have. I should have started here.
It is kind of stunning how thorough and thoughtful you are in all of your videos.
Audio has always been the hardest thing in setting up my stream. Thank you for going into audio and explaining it where its easily understandable!
Biggest thing for me soundwise was adding a compressor on my system and chat sounddevices for my GoXLR with EAPO. I can't live without it anymore. Everytime I see people struggling with loud games etc. I tell them this. Lifesaver.
The OBS color markers always annoyed me because it causes a lot of people to think their mics are peaking when they go into the red. I wish that "true peak" was set by default, because it sets the red zone around -3 dB and is much more accurate.
By the way, I love the way you showed to set initial gain. I already did most of this (like setting the compressor and then a limiter after it) as well as compressing other audio based on when I speak. But it's cool to see the whole thing. I do hope that this video gets popular because SO many people on Twitch have inconsistent audio and it hurts to watch.
Edit: My only critique, personally, is where you set your levels for background and game music, especially using the original OBS marks. Back in September of 2021, someone published "Recommendations for Loudness of Internet Audio Streaming and On-Demand Distribution". There's a few things in there, but the main thing is that it recommends most audio to be normalized to about -18 LUFS. This document inspired me (despite lacking normalization on OBS) to try to at least get the audio closer, so I generally don't allow my minimum audio to really be lower than -25 or -30 dB.
So in my case, I often play music around -20 to -30 dB depending on the song or what we're doing in stream, game audio often sits in the lower part of yellow (which would be about -12 to -20 but is often just normal volume and can reach -5 or so), and then my voice usually sits around -5 to -12 dB. On top of that, I run ducked compressors on my music and game audio, only slightly though. This allows my voice to be easily heard, but for people to still get most or all of the original sound as well.
What this also does is, mainly, it helps people that watch on TV or Phone. Often times, with most streams, the audio is so uncompressed that the music level compared to headphones is so low, that when someone talks or gets excited, my speakers are getting blown out. As far as how this translates to headphones, I actually think it helps a lot, especially for people that use it as background noise. The audio is much more "normalized" kind of (mainly consistent), so they can set a volume and not worry about peaks or it getting too quiet. Though that's obviously my own opinion, and the goal is that they are still separated, but much closer together.
Edit 2: Ah, I guess you basically do that at the end. My bad, lol. Though, I do think it would've been nice to just suggest that everyone basically does ducking, and offer non-ducking as an alternate solution where you probably mention "true peak" and the levels there.
Edit 3: One tip for anyone using a GoXLR, I prefer pulling in the music directly from the input, and not from any outputs. This bypasses the GoXLR and allows you to set your music volume independently. This gives you the benefit of ALWAYS having your music at the same level no matter what (I also have a streamdeck button to set it to 80% in Spotify), and you can then compress it slightly for normalization. It's very handy.
This is easily one of the best videos around when setting up both mic and gaming audio.
I had issues with my game audio either being too loud with my voice or my voice booming over the game audio and finding out Ducking is awesome.
Thanks, awesome video
THANK YOU. i remember wanting to find a video like this months ago, however many videos didnt exactly apply to a stream/gaming environment. you really hit the nail with the info i was looking for back then.
Hi all! I’ll skip right to it, wow does this work! Amazing settings. The only problem, and it’s not with this guide but with games, is that games have different base volumes. My mic maxes around 15-20mb which sounds good. But shooters are particularly loud. The dampening while speaking is a godsend. Without speaking, your game volume should max at 28-30mb. This was around -30 on the slider for bf4. Jedi survivor would require around -22 to reach the 30mb cap. Shooters are just particular loud. Hope that helps!
I love audio ducking, it's a game changer I use it in all my videos and streams. This is a great video
Man! You explain things that nobody I heard explains. You really did an amazing job bro!!!! I wish I saw this a while back before buying equipment. Thanks for the info!
This is the best video on RUclips covering the topic of stream audio in OBS.
This is an excellent video. So many of the "stream gurus" of yesteryear silently moved over to shilling sponsored gear which makes it impossible to find objective opinions. You're teaching critical stuff like gainstaging which streamers never find online because they think channels like podcastage are for audiophiles only. Subscribed and look forward to more :)
I do not know if anyone else had this. But the closed captioning went haywire at the halfway point. The first section was great, I am pretty sure you actually took the time to edit the CC. It looks like the time codes just ran off. It was after an add, and I am watching on an Android phone. Great video. Keep up the great work
This is the best audio guide i've seen, great job. Now my $30 Fifine sound awesome in OBS but would this work with something like EqualizerAPO so my comms on discord, zoom and in game chat benefits from it? (just the mic filters)
yes. equalizer apo is the man
Dude! Thank you so much for this video. I had major audio issues when I first started streaming. Your content is always so helpful ✨
Has a very new streamer, who has a very basic understanding of audio, but really do want the best for my viewers in the live setting, and for produced content, this is amazing. Thank you so much!
This video should be the first one you should watch on the entire channel. Great content sir.
Great video was having issues with my streams lately and this will help a bunch o7
Dude this is a splendid video! As someone who has seen many videos on this topic, I can say that this was by far the most helpful, accurate and instructional of them all. Congrats and thanks!!
9:45 The different filters, gains and so on in all the voice chat software annoyed me. I stumbled upon "Virtual Audio Cable" (VAC) (free version for non-commerical use). In OBS you select the VAC as monitoring device. You can then send your audio to output AND monitoring. Your video conference software can then use the VAC as input device.
Thus all your finely tuned OBS settings can be used within every voice chat with the same gain, noise suppression and so on.
I've been using that because I wanted to send Background music to my zoom meeting without having to be sharing anything all the time
8:40 For those confused. He meant to say 50 to 60 ms :)
This is amazing! Best tutorial and knowledgeable person I found. Been looking for an in-depth tutorial and informative tutorial like this.
Superb work. Thank you. I’m having an issue with monitoring the mic audio, getting a slight delay in my headphones. I’d love to see your analysis of how to fix those issues.
Got recommended this video on discord, really helped me out, thank you!
Fun fact: you can use Equalizer APO to add VSTs system-wide so you don't need filters in OBS, yet have the same functionality, while equally making it work in other programs.
Downside is that you don't have a visual of how these VSTs are applied, so if you need that you can test out settings in OBS first (or record and check in Audacity after each tweak) to then apply it.
This is great!
This video is so fucking complete, my god.
This saved not only my stream volume but also my video audio!! Thanks so much!
I'm going to be working on this on weekend. I always want my Audio to be perfect.
Such great info in here! Definitely gonna try setting up the sidechaining.
Really, really good video. I love the insight into your new processing chain and when the gains come in, as well as the step by step on auto-ducking. Will definitely be coming back to this one!
I know this video was posted a whole back but can I say THANK YOU! I finally was able to make my audio less shitty thanks to you ❤❤❤❤
“Home Improvement” style sound response at the beginning? Take my like and share!
Great vid, thanks! My OBS blew up & I needed to redo all my settings but its been a while since I set everything up so I needed a refresher.
fantastic tutorial - to the point, simple, well explained thank you
I used to use the OBS compressor until I got a DBX 286s. . . entirely different animal. Good video EV...
Awesome video! Thanks for the help EposVox!
This video can also help you guys with music production😉 This translates very well he's essentially teaching how to mix and its a very good tutorial
great storytelling on this one epos gg brother gg
I love this! Having studied audio engineering it always confuses me how people have their gain so high when they can fix it all in post and have plenty of control I have always ran my sound through a DAW especially with the GoXLR as the ASIO works really well
so you pass it through the DAW and then to OBS?
@@PurpleSpiritFoxFire not entirely, I use Reaper to process my sound as I’ve spent hundreds on plugins so I may as well use them. Then I route it to the Sample channel on my GoXLR mini then on the routing table for chat mic I have it ticked on only sample. Then it just mixes with my broadcast mix on stream. I still have fader control on my mic and muting, even bleeping works still. All with next to no latency. I’m sure Addie could explain in further detail if the interest was there!
By sound I mean Mic only btw. Don’t want to confuse anyone. All the other routing is done normally through the go xlr app
@@Totallyquantumofficial cause i also have plugins but using them in OBS is buggy and doesn't quite work.
@@PurpleSpiritFoxFiredo you have a GoXLR? If so the process Is fairly simple, if not it is doable but you can run into audio issues using virtual audio cables etc
Amazing video! Definitely gonna jump back into our audio settings using this as a guide!
This is (I think) going to help me a lot.
I sing to backing tracks through an analogue usb mixer into OBS and I tried to add filters to my mic, but when I played the backing track, it was all choppy and impossible to sing along to.
Obviously because the filters had affected the whole signal from the mixer.
Thanks for this, I'm off to give it a shot lol 😉👍
Great video all around !
Amazing video buddy, really in-depth but easily explained and understood. Legend as always.
Audio ducking is very cool. Thanks Epos!
Very, very, very cool video. I've been doing YT vids as a hobby for years and had already learned just about all of these, but that tooks months of mistakes and learning the hard way. One thing i really look forward to about this video, is OMG MAYBE NOW WE'LL HAVE LESS BG MUSIC DROWNING OUT THE PERSON TALKING! gd that's a pet peeve of mine. You even mentioned mic technique at the end...though you put it after the outro...that's bad because i think people are gonna click off of the video thinking it was over and that the end would only have "subscribe to me" crap at the end.
I mentioned mic technique at the end because it's an entirely separate video
@@EposVox i get that, but i was trying to say people probably click off the instant they see the outro graphic - thinking there's no more information otherthan typical YT'ber "hit like and subscribe" plugs.
Saved under favorites for later review, next morning in going to use this in obs ^^.
I love all your videos. For videocalls (zoom, g2m) to avoid the low voice volume I'm using all the knowledge I've learnt here as I'm sending OBS audio via a Virtual Cable with Audio Monitor plug in. I don't know if that is the best way but It's working atm.
Equalizer APO seems to also be a great free open source software alternative to using Voicemeeter/Steelseries Sonar, thought I'd mention it just in case
It’s amazing as it it sets audio for everything not just a single program.
Alright, time for me to go back and lose my mind tweaking audio settings again! (I kinda love it though, and this will certainly be helpful) Thank you!😂👍❤
I'm an audio engineer and producer, and I never let audio go into the red. Red always means STOP. Apparently it's fine for gamers to overdrive audio levels well into the red, and not worry about what it sounds like on the other end of the stream. Yes, audio mixes are subjective--get ten engineers and ask their opinion, you'll get twenty different opinions, but audio levels are pretty standard. I just never let audio get into the red.
I mean, objectively if you’re not going past 0dB, you’re not clipping or overdriving anything.
While what you said is generally sound advice (ha ha) that completely ignores the context of who designed the red and what it was for. Peaks being under -6db will be way too quiet for normal viewers and result in a horrible experience when they crank audio up to hear it and then click on any other video or stream.
I was wondering why my mic audio in OBS was showing just shy of red, but in the final mix I felt like it was kinda quiet. Looks like I got some re-working to do with gain and compression filters!
I didn't even know that you can add audio filters in obs before watching this video (yes, I'm dumb). Thank you sir.
Great vid as always! thanks Epos
Thank you, internet guy!
Entertaining and informative. Thanks!
If you use a GoXLR would you recommend using the GoXLR's broadcast mix for the stream's audio and ignoring the audio filters (aside from a limiter) in OBS, or would you prefer to to keep each audio track separate so you can apply more audio filters (such as ducking) for each track in OBS for the stream?
If you want to use a full mix off your interface (I'm on a WaveXLR, so I use my full mix out of Wave Link) you can still get a "ducking" effect using gain > compression > limiting. Set your relative levels of (say) music, game, mic in ascending order...in my case, music is at something like* -32dB, game at -24dB, and mic at -18dB, all in RMS (not peak!). Now, obviously your music will be far too quiet and your game won't be very loud either.
That's fine, because we'll throw like 7-8dB of gain on the whole mix.
Now your mic will be too loud. That's fine, do a 2:1 compressor with a slow attack (like 500ms) and a slow release (like 1000ms), so you don't get unnatural "pulsing." But that compressor will push down the "quieter" parts of the mix any time the "louder" parts come in. So if you're not talking but the game is going, the game audio (at 6dB above music) will push down the music. If the game gets quiet, the music will come up (as the compressor releases). And if you talk, everything gets pushed down as the compressor attacks again.
The limiter stage will prevent the gain at the front end from pushing your mic (or game) into clipping...you'll potentially get some subtle limiting on your voice (if you're being loud) for the first 500ms as the compressor attacks, but it won't be remotely noticeable, and the end result will be a nice effect where you're always front and center, but as "louder" parts peel away "quieter" parts will fade back in (with the 1000ms compressor release). All without needing access to sidechaining or a bunch of different sources in OBS.
You can also add like 2dB of gain between compressor and limiter, to boost overall loudness, if you find your total mix isn't quite loud enough.
You can actually get all of this rolled into one plugin, too. I personally use TDR's Limiter 6 GE, but that does cost money...it ran me about $30 on sale. You can also just as easily pick up (from the same TDR website) the original VladG Limiter 6 plugin, which does all of the above, only functions you'll lose are saved presets (outside of OBS) and the loudness monitoring on the final output stage. That and active support, obviously, because it's an obsolete plugin. It does work though, I've used it.
Free and paid versions, respectively:
www.tokyodawn.net/vladg-limiter-n6/
www.tokyodawn.net/tdr-limiter6-ge/
These plugins are the main reason I hate OBS's lack of a "final mix" that you can add plugins to, because a plugin like this on your final mix lets you achieve some really nice effects like I described above, and is great for managing your overall loudness levels and dynamic range, but kinda requires you to have everything all muxed into one final stream to apply it to. But for those with devices like GoXLR, WaveXLR, etc. this is a very nice option for mixing everything outside OBS and just applying one last filter to bring it all together.
* - These measurements are actually in LUFS, not dB...LUFS are very similar to RMS dB, but not quite the same.
I personally have my chat as broadcast (because you can't use chat mic) and then system on chat mic, then music is pulled directly from the GoXLR input rather than any outputs. This allows me to set music in OBS completely separately, as well as apply chat and system sounds with different filters.
@Well Well Well >no matter what you do over time ... audio will desync
Er no it won't. Maybe some of your audio or video is in 59.94 and some is in 60 FPS?
Very helpful!
I use EqualizerAPO to apply all of my mic filters incl. compressor, limiter, and sometimes RNNoise(using vst thing) because laptop fan, to the microphone device directly, and all apps get the processed audio, without using virtual audio cable or another sound device or whatever. It's kinda spotty with realtek drivers though, I had to use generic microsoft drivers to make it work, but it works well when it does.
rnnoise is godsend
What should level should the windows microphone be at? I feel like this is common knowledge but no one ever addresses it and I want to know for my own sanity. Great video as always.
thanks for the video, there is a tiny mistake , when you are talking about the limiter, you say that its good to have the release at 60 db instead of 60 ms. xD
I've been having the problem in games like Apex, or which involve gunfire at all, with that sound peaking really badly and being too boomy. Do you have a suggestion for what to set up exactly to keep the game audio evened out? Your Apex clip sounded pretty good
Needed 60dB of gain on Presonus IO24 for Dynacaster to hit average peaks between negative -18 and -12dB. So much for turning my gain down (it's maxed out). Great video though. P.S - The quality of your videos has really gone through the roof as of late!
I think I had a fundamental misunderstanding of filter/signal chains. So... if your input after hardware gain is above 0, that audio is pretty much ruined and unrecoverable. Okay. But if I'm understanding this right, you can use a gain filter that would bring the audio above 0, and as long as you get it back below zero with compression and/or a limiter... it won't be destroyed? It's not destroyed before it reaches that point in the filter chain?
If I'm understanding this right, this will merit completely reworking my filters. I've always put my additional "boost back" gain LAST in my chain, after compression/limiter, because I didn't want to clip early in the chain... but it sounds like I don't need to do that, and should put it first?
I've been using an OBS plugin by bozbez called "Win-Audio-Capture" and it's allowed me to set each individual exe file with a seperate audio capture without using Voicemeeter. That way I can tune everything individually through OBS. So before my friends in Discord sounded quiet to me but for Stream they were really loud. Now I just added an audio capture source targeting discord exe and I've turned them down on OBS.
Awesome tutorial, thanks!
God damn I love this channel... subbed!
Thank you!
Great tutorial, Thanks. I can see that you put the order Boost Back -> Compressor -> Limiter. If I have to add EQ, where should it be put in the sequence?
Thank you for this, got my sub. Gonna go do this now
I made a cheep 30 buck mick sound good so far. the side docking is the one part that I have tried to get right and no luck so far.
Do you have any recommendation for gaming headset with microphone? I have gaming headset Corsair HS80 ... my sound always on streams was fine but I notice that time to time, when I laugh loud or say something very loud it was going in that red line. Now last days it was getting even worse. I'm already using RTX voice for suppress the sounds around (background sounds) and in OBS it was now going in that red line. I literally followed up to your video and done everything the same, however where is BOOST BACK - instead of putting 16db ... I put -6.00db. and its sounds better than if I put 16db. Do you know why for me its sounds better with minus 6db instead of positive 16db? Because I put positive 16db it will be clipping and going into that red line. Do you know why?
Great video Epos, informative as ever! i'm struggling though in so many areas with my sound it's hard to know where to start first. Does this tutorial apply to condenser mics aswell? i have the yeti and i followed the section on bringing levels back up by adding gain filter set at levels to counteract peak levels lost (from compressor) as u suggested.Although i've added a limiter to avoid clipping, it sounds very blown out and really unpleasant on the ears. One question i always have is, do we set our mic inputs on windows to be 100 and control volume on a software/mic side (my mic has a physical gain ) or do we set input on windows relatively low and boost back up in obs with gain filters etc? or does it not matter which way you approach that? apologies for 20 questions, i'm never saitisfied with this damn mic ha! Great video though, very thorough as it has already been stated. You always come across very sincere and passionate about spreading the right information to the ppl that need it.
This applies to all mics.
Right click that 100% slider in Windows and change it to dB
If you’re using a XLR mic with an interface or a USB mic with its own gain dial, you want that at 0dB (which could be 100%, 50% or any other number sadly)
If this is your only place to change gain, set gain here
@@EposVox Thanks for the prompt response. i've gone back to 50% on windows side as it's the only way i can kinda clean up the sound. i guess its abit of a balancing act huh. Thank you though, i plan to invest in a dynamic/xlr set up at some point in the future with an interface so hopefully i can yield better results there.
@@EposVoxayo Epos you pick and choose which comments to reply too bozo
@@spenddollarzent ??? Yes? I'm only one human and only have so much time to reply to comments. I reply to all the ones I can/I think would be productive
@@EposVox excuses I left a comment a year ago and you responded to comments that Commented after mines
awesome video! thank you for so many helpful information. :)
I have access to the following equipment:
DBX 286s (analog audio strip)
Rode Procaster Pro (v1)
Wave XLR
Computer running Reaper (FX Chain)
A good XLR mic (SM7B and RE20)
I also have a Zoom F6 because I wanted the ability to record 32-bit float in the field, but it can double as an audio interface too (nice quiet preamps). I thought about connecting the DBX to this, but wasn't sure if that would negate the anti-clipping superpowers of the F6.
I'm not trying to stream gaming or anything. Just content production in general. Is there a significant difference between analog filters like in the DBX or software based (the others). I believe I can have a copy of the audio recorded to SD card in the Procaster pre-fader (pre-fx?) as a backup, then apply fx manually in Reaper in post. I believe the Procaster also processes the fx on the device before sending to the computer to OBS or a recording app - this might be the ideal streaming setup in my case.
Basically, I have a computer for production (recording & editing, not streaming) and a computer for streaming.
Hardware better than software. Less latency, which is important if you’re routing things to different places. Ideally you should consolidate to a high end interface that has DSP like that built in. If you’re more of an audiophile, you would then want to add a preamp even in front of the interface.
Followed this to the letter and my god you are a saint. I always wonder about my audio in streams and after testing all of this, I definitely have good headroom and better audio quality.
Just to add, the ducking effect IS lowering the Broadcast Stream Mix in OBS, but my mic drops off to nothing or doesn't get recorded when I have that sidechain compressor enabled. Maybe I'm trying to duck something that shouldn't be ducked?
thank you very much 🙏
Thanks!
Keep in mind the lower your peak recording gain the more you are going to have to add gain up in post. This will make your room noise louder. You may start to hear your ac and untreated room more.
This isn’t true. Your background noise scales down with your gain if you do it right
@@EposVox Maybe I expand my self poorly. It is true tho. Anytime you turn up a recording in post you are also turning up the room noise in that recording. If I have a recording where my voice peaks at say -12 db and I turn it up digitally in post it will also raise the room noise when im not speaking. Anytime you are adding gain you are raising your noise floor with your main signal.
Ok, but if you turn up gain on your interface instead, that same room noise comes up - plus the risk of added amp noise. Gain in post doesn't bring it up anymore than your amp gain would.
9:48 There is away to run OBS into other apps with a Virtuial Audio Cable, Voicemeter lets u download one for free and i set it up so i can run OBS into discord and VRChat so my mic doesn't sound dull or too quiet.
Great overview, but am I tripping or are most big streamers much louder on their mic? Seems like everyone must be within the red most of the time.
They often are; that’s not good practice
I personally use minimal compression and a limiter (-15db atm) so that if I do get loud it keep that in check.
Limiting to -15dB would be WAY too quiet compared to every other stream lol
@EposVox The compressor plus mic gain bring it up around -8 on the output. I have voicemeeter running it.
How is the microphone that you recording on transfer over to in game chat?
Awesome video.
04:49 - 05:37 this is the most replayed part (for now)
Then why should we set preamps to 75% and not 50%?
I set my gain to the level where my noise floor is around -60dB (less than 3 o'clock)
And it never hits -6dB 01:37 because I talk very quietly. Most of the time it would be around -27dB
02:17 - 02:24 It would be around 100% gain for me.
Is there any way to automatically balance two audio sources to the same volume?
Is this something you can do fully in something like the Wave:3 software?
What should my Compressor & Limiter be at for voice chat?
Also the EQ & Compressor for Game?
When watching this I think you're confusing Dynamic with Condenser, you say your gain dial will be lower if using a dynamic mic which isn't true, condensers have way more sensitivity so require less gain where dynamic generally needs more since they're mics that can take a beating, like on stage for example. I know this video is old but that point really confused me. To reach the points you want without using any OBS my dial is already at 75%