Sapphire
11-04-2009, 06:39 PM
Whether you're a singer who's trying to mix his/her vocals into an instrumental track, or a producer who's trying to clean up the vocal tracks for his/her project, there are things that need to (or can be) done to your vocals to make them sound crisp and clean.
This tutorial will teach you how to clean up vocal tracks so that they not only stand out nicely, but also sound a little natural and high quality. To be able to do this, there are a few things you need to get familiar with:
Basic Filters and Effects:
Noise Removal
If you don't use this, then you're crazy or you just aren't familiar with it. Noise Removal is used to reduce or remove any background, humming, hissing sounds within your audio. Sometimes it works 100%, but other times, if the noise is just too much, Noise Removal will make your audio sound very tinny.
My favorite program to use this feature in is Goldwave. It's not free if you're ignorant (about $25-30), but I think it's worth it for recording and cleaning purposes. Adobe Audition/Cool Edit's Noise Removal feature is so-so. It'll reduce the noise, but not completely remove it, still BGM can mask any left over noise. Audacity has a Noise Reduction/Removal feature, but it's not the best. Most of the time your audio can come out tinny, but if you're an expert at Audacity then you probably don't have this problem.
If you're using Adobe Audition or Cool Edit, and want to learn how to use the Noise Removal feature, then check out this lovely tutorial (http://voiceactingalliance.com/board/showthread.php?t=7520) by Tamtu. Even if you don't have any of the programs, you should still check it out to know what the feature does.
For those of you who use Goldwave to record, here is a mini tutorial on how to use the Noise Reduction feature. You'll want to download this file if you plan on following along (or you can use your own recordings (http://www.4shared.com/account/file/146309744/b9d6a52f/VoiceTest.html).
*Thanks to Y. Chang for letting me use his recording as an example!
1. Open up your audio of choice (preferably, one with the background/static/hissing/humming interference). Like so:
http://img14.imageshack.us/img14/5996/noiser01.png
2. Ugh, listen to all that noise. Let's get rid of it by selecting a portion of the noise from either the beginning of the recording or the end of it (wherever you hear the noise without the vocals is what you want to select). Then you're going to RIGHT CLICK > COPY the selection. Hit Ctrl+A to select the entire waveform.
http://img690.imageshack.us/img690/4739/noiser02.png
3. Then go to EFFECTS > FILTERS > NOISE REDUCTION and under 'Reduction Envelope' click the bubble that reads 'USE CLIPBOARD'. Click 'OK'
http://img14.imageshack.us/img14/865/noiser03.png
4. Playback the file and you'll notice the background noise is gone! It should sound something like this: End Product (http://www.4shared.com/account/file/146320687/8a60cd44/VoiceTest__Complete_.html[/url)
http://img203.imageshack.us/img203/3840/noiser04s.png
For those of you who use Audacity-- you shouldn't. BUT if you must, then I recommend checking out this tutorial on proper Noise Removal. (http://wiki.audacityteam.org/index.php?title=Noise_Removal) Provided by makers of Audacity.
------------------------------------------------
Compression
What is this!? Well, to be honest, I can't really explain the technical crap behind this filter, but the one thing it does is make audio stand out. There are a number of filters that can do that, but compressors are what make decent audio sound professional. A lot of artists, sound engineers, etc use compression for their music, especially on vocal tracks. I'm not going to go into too much detail about this feature because it's one of those things that are done based on taste.
You usually want to do this after you've noise reduced/removed your vocal tracks. Otherwise, when you add a compressor to something with noise in the background, it'll not only make the vocals stand out, but it'll make the noise stand out too and we don't want that. Also, when you use a compressor USE IT ON SIMILAR TRACKS, ONLY. For example: You finished mixing in the vocals, bgm, and sfx for your fandub, and you've already saved them all into one big file. DON'T use a compressor on that file. Instead, you should only use it on vocals or individual instruments (if you're making a song). Otherwise, I think it'll sound weird. :/
Another thing: If you have a low quality mic, this may benefit you, but if there are any noticeable room echoes in your recording BEFORE you add the compressor, then the room echoes will stand out after you add the compressor. I like to add a little Reverb to low quality recordings after I add a compressor to low quality recordings.
Now, unfortunately, unless you have a compressor plugin, this is not available for Audacity and Goldwave. With that said, if you have Adobe Audition v3 or earlier, Cool Edit Pro, or Adobe Soundbooth, then this will apply to you, but I'm using Adobe Audition for this tutorial.
1. Let's open up the 'Voice Test (Complete)' file that was included in the Noise Removal tutorial above.
http://img217.imageshack.us/img217/9504/comp01.png
2. Go to EFFECTS > AMPLITUDE AND COMPRESSION > TUBE-MODELED COMPRESSOR. Some people use Multiband Compressors but I don't think it does as good of a job as Tube-Modeled does.
http://img691.imageshack.us/img691/8792/comp02.png
3. Now, look at that. This looks a little simple, right? Good, because it really is self-explanatory. Usually I just mess around with the presets and then if I find a preset that sounds close to what I like, I'll use it and adjust the settings at the bottom left of the window. I think it'd be safe to just mess with the Threshold slider. Most people keep the Attack at 0.1 ms. You should too, but it never hurts to mess around with it. The Gain slider reduces or increases the volume of the compressor. Nothing more. Really, just play around with it. This is what the audio sounds like with the settings below: Voice Test w/ Compression (http://www.4shared.com/file/146359920/2a9f64ca/VoiceTest__Complete-COMP_.html[/url)
http://img204.imageshack.us/img204/4347/comp03n.png
CoolEdit Pro and Adobe Soundbooth have compressor features, so I would look around for it and just play around with the settings until you get something you like. C:
Next Time: Reverb, Volume, and Mixing Vocals into BGM/SFX
-------------------------------------------
Question/Answer Session
I was wondering, when you mix people's lines, do they naturally have a good mic or do you add another effect to make it sound especially crisp and consistent throughout?
When I cast, I try to get decent voice actors with decent quality recordings. Obviously, everyone's mic quality ends up sounding different, which is why I add slight compressors to the low quality mics, then I'll add a slight reverb to the vocal track. In the end,it all depends on the person's quality.
Eliminating peaking. I'm sure this is an individual problem since while mixers can lower the volume of lines that peak over the mic, they can't get rid of that... "peakness", can they? Is there a way?
Honestly, for the sake of your project, I would just have the voice actor redo the line. There really is no way of removing distortion due to peaking, at least not to my knowledge. The best way to avoid peaking is by reducing your mic volume or computer recording volume (if you're a Windows User, you can do that by going into your Control Panel > Hardware and Sound) before you record.
My own mic has good quality but low volume. When I try to make it louder, the static/white noise goes up too. Since Audacity can't lower ambient noise, I need suggestions for a (free) program that can.
This is what the Noise Removal/Reduction filter is for. The static should be loud when you up the volume (it happens with mine too and I take care of it with Noise Reduction). If that doesn't work, then I really don't know what to tell you. I don't know of any free programs that can use Noise Reduction/Removal. :/
---------------------------------
Reverb
Many artists add reverb to vocal tracks for their songs. Sometimes it's a lot and sometimes just a pinch is added to it. Reverb is not only used for music purposes, but it can be used for vocals in fandubs/radioplays too. I won't go into detail about the reverb settings in Adobe Audition, but I WILL tell you how you can use reverb to make the vocals in your project kind of blend in with a background track. Reverb not only helps blend in your vocals, but it also helps them stand out and sound crisp and clear.
I'll use my latest Gurren Lagann fandub clip as a perfect example (just click here). (http://www.vimeo.com/8260948)
If you haven't noticed it yet, I added reverb to the entire vocal track which is why the voices crisp, clear, and fitting in this scene. Of course I also added compression and did a bit of noise removal when it was required, so I wouldn't rely on reverb alone.
Anyway, the reverb filters that I use the most are Studio Reverb and Full Reverb. I love them so much. You're free to mess around with the other filters, but these are my top two choices.
Here's a sample of the vocal track (w/reverb): Here (http://www.4shared.com/file/196262945/cc570c66/GLTUT_Reverb_.html
)
There's a good amount of reverb in there. Surprising, no? Yes? Whatever. Anyway, I used Full Reverb for this baby, so go to EFECTS > REVERB > FULL REVERB.
I'm not going to tell you what settings I use since it really depends on the kind of atmosphere you're going for in your project, but I'll tell you what sliders I mess around with the most. Take a look at the image below:
http://img706.imageshack.us/img706/889/reverbtut.png
The sections that are highlighted in yellow are the ones I play around with the most to get the reverb I want.
Pre-Delay/Decay: I can't really describe these two sections, but they go hand in hand with each other. These two allow you to adjust the amount of echo in your audio. You might mess around with the sliders if you're dubbing a scene that requires the characters to be outside or inside a cave or something.
Perception: Do you want sound to clash together (reflective) or do you want it nice and clean (absorbency)? This is the section to play around with. You could probably use this for things like someone's inside thoughts.
Room Size: Do I really need to explain this? Move the slider towards the left side to create reverb in a small room. Move it to the right side to create reverb in a large room/area. Since the Gurren Lagann clip took place outside (a large space), I decided to move the slider all the way to the right.
Dry vs. Wet: My favorite section. Do you want to create a dry sound for your vocals but with a tiny bit of reverb (wet)? Do this by making the ratio of Dry to Wet 3:1. If you want to do the exact opposite and drown your voice with reverb, then just reverse the ratio of Dry to Wet. You will always use this section. Trust me.
If you have trouble making your own default settings from scratch, then try using the presets already given to you in the drop down menu at the top of the window. From there, you can play around with the settings until you get something you like.
Room Reverb
Sometimes it's roomy ambiance isn't enough to create the atmosphere in your project. Maybe you want to take the realism a step further? Here are the settings I use to create a roomy reverb for vocals. You can use it for other sound effects, but you may need to tweak the settings a bit.
http://img3.imageshack.us/img3/2130/roomreverb.png
These are settings for a small room. You can increase the room size but you may have to increase/decrease the early reflections and wet settings. Make sure dry is always at a very high percentage. Delay's percentage should be much greater than Pre-delay's, etc.
------------------------------------------------
Compression (Bad/Good?)
So when do you know when to use the right amount of compression on your audio anyway? Sometimes people like to slap some compression onto a 100% mixed audio, sometimes they just add to it vocals and leave things like Music, SFX untouched. In fact, an of these ways are fine IF you know what the hell you're doing.
Remember this effect pattern:
EQ
Compression
Everything else
And you should be fine. Here's an example of what a sample audio would sound like if you put effects in the following order. Let's replace "Everything else" with Reverb for this section.
Good Compressing (Vocals ONLY) (http://www.box.net/shared/v6sme2h514)
Good Compressing (Final Mix) (http://www.box.net/shared/17pm8fm366)
Here's what it sounds like if "Everything else" came before Compressing:
Bad Compressing (Vocals ONLY) (http://www.box.net/shared/xmujcat685)
Bad Compressing (Final Mix) (http://www.box.net/shared/gn0igzf506)
It sounds... awful. Completely muddy. The reverb is way too noticeable and choppy. In the second sample with the sound effects, you can tell that everything just meshed together making the sound to crowded. You can create this mess by just reducing the threshold by a bunch of dB's in whatever compression plugin you happen to be using.
Yeah, so uh, don't do that. You think the latter method sounds pro but it really sounds like shit.
This tutorial will teach you how to clean up vocal tracks so that they not only stand out nicely, but also sound a little natural and high quality. To be able to do this, there are a few things you need to get familiar with:
Basic Filters and Effects:
Noise Removal
If you don't use this, then you're crazy or you just aren't familiar with it. Noise Removal is used to reduce or remove any background, humming, hissing sounds within your audio. Sometimes it works 100%, but other times, if the noise is just too much, Noise Removal will make your audio sound very tinny.
My favorite program to use this feature in is Goldwave. It's not free if you're ignorant (about $25-30), but I think it's worth it for recording and cleaning purposes. Adobe Audition/Cool Edit's Noise Removal feature is so-so. It'll reduce the noise, but not completely remove it, still BGM can mask any left over noise. Audacity has a Noise Reduction/Removal feature, but it's not the best. Most of the time your audio can come out tinny, but if you're an expert at Audacity then you probably don't have this problem.
If you're using Adobe Audition or Cool Edit, and want to learn how to use the Noise Removal feature, then check out this lovely tutorial (http://voiceactingalliance.com/board/showthread.php?t=7520) by Tamtu. Even if you don't have any of the programs, you should still check it out to know what the feature does.
For those of you who use Goldwave to record, here is a mini tutorial on how to use the Noise Reduction feature. You'll want to download this file if you plan on following along (or you can use your own recordings (http://www.4shared.com/account/file/146309744/b9d6a52f/VoiceTest.html).
*Thanks to Y. Chang for letting me use his recording as an example!
1. Open up your audio of choice (preferably, one with the background/static/hissing/humming interference). Like so:
http://img14.imageshack.us/img14/5996/noiser01.png
2. Ugh, listen to all that noise. Let's get rid of it by selecting a portion of the noise from either the beginning of the recording or the end of it (wherever you hear the noise without the vocals is what you want to select). Then you're going to RIGHT CLICK > COPY the selection. Hit Ctrl+A to select the entire waveform.
http://img690.imageshack.us/img690/4739/noiser02.png
3. Then go to EFFECTS > FILTERS > NOISE REDUCTION and under 'Reduction Envelope' click the bubble that reads 'USE CLIPBOARD'. Click 'OK'
http://img14.imageshack.us/img14/865/noiser03.png
4. Playback the file and you'll notice the background noise is gone! It should sound something like this: End Product (http://www.4shared.com/account/file/146320687/8a60cd44/VoiceTest__Complete_.html[/url)
http://img203.imageshack.us/img203/3840/noiser04s.png
For those of you who use Audacity-- you shouldn't. BUT if you must, then I recommend checking out this tutorial on proper Noise Removal. (http://wiki.audacityteam.org/index.php?title=Noise_Removal) Provided by makers of Audacity.
------------------------------------------------
Compression
What is this!? Well, to be honest, I can't really explain the technical crap behind this filter, but the one thing it does is make audio stand out. There are a number of filters that can do that, but compressors are what make decent audio sound professional. A lot of artists, sound engineers, etc use compression for their music, especially on vocal tracks. I'm not going to go into too much detail about this feature because it's one of those things that are done based on taste.
You usually want to do this after you've noise reduced/removed your vocal tracks. Otherwise, when you add a compressor to something with noise in the background, it'll not only make the vocals stand out, but it'll make the noise stand out too and we don't want that. Also, when you use a compressor USE IT ON SIMILAR TRACKS, ONLY. For example: You finished mixing in the vocals, bgm, and sfx for your fandub, and you've already saved them all into one big file. DON'T use a compressor on that file. Instead, you should only use it on vocals or individual instruments (if you're making a song). Otherwise, I think it'll sound weird. :/
Another thing: If you have a low quality mic, this may benefit you, but if there are any noticeable room echoes in your recording BEFORE you add the compressor, then the room echoes will stand out after you add the compressor. I like to add a little Reverb to low quality recordings after I add a compressor to low quality recordings.
Now, unfortunately, unless you have a compressor plugin, this is not available for Audacity and Goldwave. With that said, if you have Adobe Audition v3 or earlier, Cool Edit Pro, or Adobe Soundbooth, then this will apply to you, but I'm using Adobe Audition for this tutorial.
1. Let's open up the 'Voice Test (Complete)' file that was included in the Noise Removal tutorial above.
http://img217.imageshack.us/img217/9504/comp01.png
2. Go to EFFECTS > AMPLITUDE AND COMPRESSION > TUBE-MODELED COMPRESSOR. Some people use Multiband Compressors but I don't think it does as good of a job as Tube-Modeled does.
http://img691.imageshack.us/img691/8792/comp02.png
3. Now, look at that. This looks a little simple, right? Good, because it really is self-explanatory. Usually I just mess around with the presets and then if I find a preset that sounds close to what I like, I'll use it and adjust the settings at the bottom left of the window. I think it'd be safe to just mess with the Threshold slider. Most people keep the Attack at 0.1 ms. You should too, but it never hurts to mess around with it. The Gain slider reduces or increases the volume of the compressor. Nothing more. Really, just play around with it. This is what the audio sounds like with the settings below: Voice Test w/ Compression (http://www.4shared.com/file/146359920/2a9f64ca/VoiceTest__Complete-COMP_.html[/url)
http://img204.imageshack.us/img204/4347/comp03n.png
CoolEdit Pro and Adobe Soundbooth have compressor features, so I would look around for it and just play around with the settings until you get something you like. C:
Next Time: Reverb, Volume, and Mixing Vocals into BGM/SFX
-------------------------------------------
Question/Answer Session
I was wondering, when you mix people's lines, do they naturally have a good mic or do you add another effect to make it sound especially crisp and consistent throughout?
When I cast, I try to get decent voice actors with decent quality recordings. Obviously, everyone's mic quality ends up sounding different, which is why I add slight compressors to the low quality mics, then I'll add a slight reverb to the vocal track. In the end,it all depends on the person's quality.
Eliminating peaking. I'm sure this is an individual problem since while mixers can lower the volume of lines that peak over the mic, they can't get rid of that... "peakness", can they? Is there a way?
Honestly, for the sake of your project, I would just have the voice actor redo the line. There really is no way of removing distortion due to peaking, at least not to my knowledge. The best way to avoid peaking is by reducing your mic volume or computer recording volume (if you're a Windows User, you can do that by going into your Control Panel > Hardware and Sound) before you record.
My own mic has good quality but low volume. When I try to make it louder, the static/white noise goes up too. Since Audacity can't lower ambient noise, I need suggestions for a (free) program that can.
This is what the Noise Removal/Reduction filter is for. The static should be loud when you up the volume (it happens with mine too and I take care of it with Noise Reduction). If that doesn't work, then I really don't know what to tell you. I don't know of any free programs that can use Noise Reduction/Removal. :/
---------------------------------
Reverb
Many artists add reverb to vocal tracks for their songs. Sometimes it's a lot and sometimes just a pinch is added to it. Reverb is not only used for music purposes, but it can be used for vocals in fandubs/radioplays too. I won't go into detail about the reverb settings in Adobe Audition, but I WILL tell you how you can use reverb to make the vocals in your project kind of blend in with a background track. Reverb not only helps blend in your vocals, but it also helps them stand out and sound crisp and clear.
I'll use my latest Gurren Lagann fandub clip as a perfect example (just click here). (http://www.vimeo.com/8260948)
If you haven't noticed it yet, I added reverb to the entire vocal track which is why the voices crisp, clear, and fitting in this scene. Of course I also added compression and did a bit of noise removal when it was required, so I wouldn't rely on reverb alone.
Anyway, the reverb filters that I use the most are Studio Reverb and Full Reverb. I love them so much. You're free to mess around with the other filters, but these are my top two choices.
Here's a sample of the vocal track (w/reverb): Here (http://www.4shared.com/file/196262945/cc570c66/GLTUT_Reverb_.html
)
There's a good amount of reverb in there. Surprising, no? Yes? Whatever. Anyway, I used Full Reverb for this baby, so go to EFECTS > REVERB > FULL REVERB.
I'm not going to tell you what settings I use since it really depends on the kind of atmosphere you're going for in your project, but I'll tell you what sliders I mess around with the most. Take a look at the image below:
http://img706.imageshack.us/img706/889/reverbtut.png
The sections that are highlighted in yellow are the ones I play around with the most to get the reverb I want.
Pre-Delay/Decay: I can't really describe these two sections, but they go hand in hand with each other. These two allow you to adjust the amount of echo in your audio. You might mess around with the sliders if you're dubbing a scene that requires the characters to be outside or inside a cave or something.
Perception: Do you want sound to clash together (reflective) or do you want it nice and clean (absorbency)? This is the section to play around with. You could probably use this for things like someone's inside thoughts.
Room Size: Do I really need to explain this? Move the slider towards the left side to create reverb in a small room. Move it to the right side to create reverb in a large room/area. Since the Gurren Lagann clip took place outside (a large space), I decided to move the slider all the way to the right.
Dry vs. Wet: My favorite section. Do you want to create a dry sound for your vocals but with a tiny bit of reverb (wet)? Do this by making the ratio of Dry to Wet 3:1. If you want to do the exact opposite and drown your voice with reverb, then just reverse the ratio of Dry to Wet. You will always use this section. Trust me.
If you have trouble making your own default settings from scratch, then try using the presets already given to you in the drop down menu at the top of the window. From there, you can play around with the settings until you get something you like.
Room Reverb
Sometimes it's roomy ambiance isn't enough to create the atmosphere in your project. Maybe you want to take the realism a step further? Here are the settings I use to create a roomy reverb for vocals. You can use it for other sound effects, but you may need to tweak the settings a bit.
http://img3.imageshack.us/img3/2130/roomreverb.png
These are settings for a small room. You can increase the room size but you may have to increase/decrease the early reflections and wet settings. Make sure dry is always at a very high percentage. Delay's percentage should be much greater than Pre-delay's, etc.
------------------------------------------------
Compression (Bad/Good?)
So when do you know when to use the right amount of compression on your audio anyway? Sometimes people like to slap some compression onto a 100% mixed audio, sometimes they just add to it vocals and leave things like Music, SFX untouched. In fact, an of these ways are fine IF you know what the hell you're doing.
Remember this effect pattern:
EQ
Compression
Everything else
And you should be fine. Here's an example of what a sample audio would sound like if you put effects in the following order. Let's replace "Everything else" with Reverb for this section.
Good Compressing (Vocals ONLY) (http://www.box.net/shared/v6sme2h514)
Good Compressing (Final Mix) (http://www.box.net/shared/17pm8fm366)
Here's what it sounds like if "Everything else" came before Compressing:
Bad Compressing (Vocals ONLY) (http://www.box.net/shared/xmujcat685)
Bad Compressing (Final Mix) (http://www.box.net/shared/gn0igzf506)
It sounds... awful. Completely muddy. The reverb is way too noticeable and choppy. In the second sample with the sound effects, you can tell that everything just meshed together making the sound to crowded. You can create this mess by just reducing the threshold by a bunch of dB's in whatever compression plugin you happen to be using.
Yeah, so uh, don't do that. You think the latter method sounds pro but it really sounds like shit.