Just wondering if it's possible to fix a live recording that picked up a poorly mic.ed singer? A few times he got a little rowdy on the mic and got distorted. I was recording with the DR-40 in a room that had their PA system running. Most of the audio came out ok, for live anyway. Just wondering if it can be hidden/fixed somehow so it's less distracting in a video. TIA.
-
Wondering which camera, gear, computer, or software to buy? Ask in our Gear Guide.
Fix bad audio recordings?
- Thread starter icba pictures
- Start date
Thanks Alcove, does that mean mission unaffordable?
Doing things as practice right now... so zero dollars and no sense.
With the Avid Fast Track USB PT/SE Audio Interface, would I be able to do anything? Are there tutorials on how to modify this sort of thing? I've found it to be somewhat intimidating so far just to use it (PT or the interface) for anything...
I am really interested in learning more about how to use it, I've been listening to movies and music more closely as of late, it's a new world for me. Trying to notice the different use of sounds, now to figure out how to get there from here. Just not sure what to do for practice. There seems to be plenty of "take videos of this" practice suggestions but no "record these sounds and edit it" ideas. Any suggestions would be appreciated.
The problem with your predicament is that fixing distorted recordings is almost impossible, and to get even a decent result you need lots of expensive gear.
You could try EQing out the offensive frequencies for a slight improvement, but make sure you don't damage the clarity of the recordings. How bad is it anyway? Care to show us? Could you replace distorted parts with others?
Check this out too: http://www.izotope.com/products/audio/rx/
You could try EQing out the offensive frequencies for a slight improvement, but make sure you don't damage the clarity of the recordings. How bad is it anyway? Care to show us? Could you replace distorted parts with others?
Check this out too: http://www.izotope.com/products/audio/rx/
Last edited:
Don't know if I'm reading it right but I understood it was the venue's PA that suffered from distortion. From what I know that software is designed for interpolating information lost through digital clipping. Repairing what may well be a good recording of analogue distortion may not give such good results. But yeah, it's great software and could well be the best bet.
1 - Distortion is almost impossible to correct under the best of circumstances.
2 - Even if it just the vocal that is distorted you have no way of separating it from the rest of the sound, so any corrections that you attempt to fix the vocal distortion will also affect the rest of the music.
3 - There are a few plug-ins that can smooth out distortion to the point where it is not totally offensive, and they are pricey.
4 - If it is really important, and you have zero budget, you can try to rewrite the audio waveform on your own. If you expand the waveform to millisecond level you will see that it is very jagged (as long as it is not completely peaked out as well). You can use the pencil tool to smooth it out. I've done this to VERY short transients with equivocal results; sometimes it works, sometimes it doesn't. It will probably take you two or three hours per second, and you have a 10% chance of going from absolutely horrible to pretty terrible.
5 - You can us EQ to subtract the most offensive frequencies to make it a little more tolerable.
6 - No matter what you do, unless it is EXTREMELY short, it will be very noticeable in the final product.
2 - Even if it just the vocal that is distorted you have no way of separating it from the rest of the sound, so any corrections that you attempt to fix the vocal distortion will also affect the rest of the music.
3 - There are a few plug-ins that can smooth out distortion to the point where it is not totally offensive, and they are pricey.
4 - If it is really important, and you have zero budget, you can try to rewrite the audio waveform on your own. If you expand the waveform to millisecond level you will see that it is very jagged (as long as it is not completely peaked out as well). You can use the pencil tool to smooth it out. I've done this to VERY short transients with equivocal results; sometimes it works, sometimes it doesn't. It will probably take you two or three hours per second, and you have a 10% chance of going from absolutely horrible to pretty terrible.
5 - You can us EQ to subtract the most offensive frequencies to make it a little more tolerable.
6 - No matter what you do, unless it is EXTREMELY short, it will be very noticeable in the final product.
When I get home I'll try to post some clips of the actual problem. It's the "born to be wild" part of the song, I think mostly just "born" though. Thanks guys, I figured it would be near impossible, and definitely not worth the cost of fixing it, but would like to see if I can make it better just for practice sake. I do have 3 audio sources from that night, 2 camera and 1 DR-40... maybe one is less noticeable than the others???
Oops, I guess it was the "get your motor running" parts.... Anyway, here's a link to some spliced takes of the singing. I put some louder parts in to let you see the overall sound. 3 different sources, you should be able to see the differing cameras and then the audio only is set to pictures from my ride home tonight, lol, after reviewing them to make the clip though, it may be user error as I don't know jack about levels or where to set what when it comes to audio. I want to learn though, so any input would be appreciated. What did I do wrong???
http://youtu.be/DrSXl730NcQ
http://youtu.be/DrSXl730NcQ
It's just an idea but what if you took out the born to be wild and mixed in a new guitar sound or some instrumental sound instead? It might be better than distortion, and the listeners might think that it was intentional, and part of this particular cover of the song.
In these situations you're better off taking the sound directly from the mix table and then setting up an extra 2 microphones directed at the crowd (if there is one). This way you can have perfect sound from the musicians.
That might work...thanks, but fortunately this is just practice for both them and me... I will be much more careful in the future with the audio recording.
Hmmmm, mix table... Well they have a PA system of some sort that does have mix controls, so I will have to check that option out next time. Thanks. Would that be best for getting a good audio track for doing a music video? I know it wouldn't be studio quality, but they don't have the funds but I'm just trying to make something nice for them to share with friends and family (I still want it to sound studio quality, lol). This band ain't trying to go pro, but I want to do the best I can for them. There is a talent show coming up that they plan to perform for, in March. I'd like to get that on video for them too, so am trying to get rid of any of these problems now.
There are number of issues that you need to address.
First, is this a performance ("concert") video or a music video; these are two VERY different things. A performance video is a live performance captured on "tape." A music video is lip-syncing/playing to a prerecorded audio track.
The ultimate "perfect" concert video will have a multi-track recording of the individual instruments and vocals that will be re-mixed after the show. This, of course, is not usually an option for club bands, etc.
The next best thing is to record the audio from the FOH (Front of House) mixing console. The problem here is that in smaller (and even much larger) venues the FOH mix is mixed so the band sounds great in the room, it is not for later reproduction. This means that in a small or even medium sized room the mix captured off of the FOH console will most probably be very unbalanced; the guitar player may be playing very loud so may be very quiet or almost non-existent in the FOH mix except for what is picked up by the other mics on stage, and that will be flukey as the singers move on and off of the vocal mics, and can get especially weird if the lead vocalist walks around the stage, holds his/her mic to the side or points it towards the audience when not singing, for example. In very small rooms it is not at all unusual for just the vocals and the kick drum to be miced, and that is all you will get from the FOH mix.
You can use one or two mics aimed at the PA cabinets and/or the stage for a better blend. However, if the mics are very close you may not get good representation from the lower end of the audio frequency spectrum (bass and kick drum, for example) as low sounds need more space to propagate. As you move further back, however, the issue becomes crowd noise. In this instance you need to find a balance between the two, a compromise that is never wholly satisfactory. (BTW, the mics of the DR-40 are woefully inadequate to the task.)
If you have more audio tracks available you can record the FOH mix direct and the PA/Stage mix with mics, then blend the two together during post. It is also very nice to have a pair of audience mics. The best I ever accomplished in a medium sized room was a mix off of the FOH console, a pair of mics above and about ten feet in front of the stage, a pair of mics directly in front of the FOH console (the best mix, as this is what the FOH engineer is hearing but with a fair amount of crowd noise) and a pair of crowd mics for a total of eight (8) audio tracks. It took a while to correct for phasing issues as they were not perfectly in sync; sound does take time to go from the front to the back of the room. When I "mixed" the show I made multiple copies of the stage and FOH audio tracks and then used EQ to bring out and even reduce certain instruments a little bit. That took almost as long as mixing for individual instruments.
Once again, this is something that takes practice and experience to get right. See if the band has a couple of audio recorders (most bands record rehearsals) and a few extra mics. Maybe their live mix engineer, if they have a regular guy/gal, can do the post audio for you.
First, is this a performance ("concert") video or a music video; these are two VERY different things. A performance video is a live performance captured on "tape." A music video is lip-syncing/playing to a prerecorded audio track.
The ultimate "perfect" concert video will have a multi-track recording of the individual instruments and vocals that will be re-mixed after the show. This, of course, is not usually an option for club bands, etc.
The next best thing is to record the audio from the FOH (Front of House) mixing console. The problem here is that in smaller (and even much larger) venues the FOH mix is mixed so the band sounds great in the room, it is not for later reproduction. This means that in a small or even medium sized room the mix captured off of the FOH console will most probably be very unbalanced; the guitar player may be playing very loud so may be very quiet or almost non-existent in the FOH mix except for what is picked up by the other mics on stage, and that will be flukey as the singers move on and off of the vocal mics, and can get especially weird if the lead vocalist walks around the stage, holds his/her mic to the side or points it towards the audience when not singing, for example. In very small rooms it is not at all unusual for just the vocals and the kick drum to be miced, and that is all you will get from the FOH mix.
You can use one or two mics aimed at the PA cabinets and/or the stage for a better blend. However, if the mics are very close you may not get good representation from the lower end of the audio frequency spectrum (bass and kick drum, for example) as low sounds need more space to propagate. As you move further back, however, the issue becomes crowd noise. In this instance you need to find a balance between the two, a compromise that is never wholly satisfactory. (BTW, the mics of the DR-40 are woefully inadequate to the task.)
If you have more audio tracks available you can record the FOH mix direct and the PA/Stage mix with mics, then blend the two together during post. It is also very nice to have a pair of audience mics. The best I ever accomplished in a medium sized room was a mix off of the FOH console, a pair of mics above and about ten feet in front of the stage, a pair of mics directly in front of the FOH console (the best mix, as this is what the FOH engineer is hearing but with a fair amount of crowd noise) and a pair of crowd mics for a total of eight (8) audio tracks. It took a while to correct for phasing issues as they were not perfectly in sync; sound does take time to go from the front to the back of the room. When I "mixed" the show I made multiple copies of the stage and FOH audio tracks and then used EQ to bring out and even reduce certain instruments a little bit. That took almost as long as mixing for individual instruments.
Once again, this is something that takes practice and experience to get right. See if the band has a couple of audio recorders (most bands record rehearsals) and a few extra mics. Maybe their live mix engineer, if they have a regular guy/gal, can do the post audio for you.
They are not even a paid band... Just a few guys jamming on a regular basis, doing a party or two each year.
They only have a mix board (PA) for the singer(s), nothing else is mic.ed. And that mixer is fairly new to them, before it was just a PA amp and speakers.
For me to get good levels from the DR-40, do I want to get things to -20db average?? Been reading all the info about gain staging, I'm trying to follow along with the details. Also should I mic the center of the "stage" just in front? That's how I did it that night but my input levels should have been lower where it was located I guess.
And, should I be recording with two identical mic.s plugged into the DR? Balanced mic.s of course??
Man I have a lot to learn.That makes it tough, all right...
The tough part is going to be finding the "sweet spot" that captures a good balance of PA and band. Surprisingly this can be almost directly over the stage - if the band is using vocal monitors. Otherwise, as I mentioned, most often it would be in front of the stage so you get the stage instruments and the vocals from the PA.
It's always a guessing game. Between -20dB and -12dB should be okay. This is one of those instances where you will want to engage the limiter, but where it cuts in a few dB above your optimum level. If you're general level during sound check is -20dB the band will be hitting -12dB by the end of the first song and be climbing from there, so set the limiter so it cuts in at about -10dB and -8 dB. You may want to start at a lower level.
As an aside, back when I was gigging clubs consistently, I almost always boosted my main level from three to about four (I had a VERY powerful system for my keyboard rig) by the end of the first song, and as the club filled and the rest of the band cranked it up I might get all the way up to six. It's funny, but the bigger the stage the less I had to increase volume over the course of a night as I was more isolated sonically from the rest of the band and relied more on the stage monitors. That's why I loved gigs like the Beacon Theatre, the Apollo, Carnagie Hall, outdoor festivals, etc. Those huge pro systems were freakin' AWESOME!!!!! They were so big and powerful you didn't have to get overly loud.
... should I be recording with two identical mics plugged into the DR? Balanced mic.s of course??
Yes, a matched pair of decent mics will sound a lot better than the DR-40 internal mics. They are also easier to place; "hang" them from the ceiling if you can - you may be able to "anchor" them to a light bar/truss.
You bet your a$$!!!!!!!!!! Don't worry about it; one of the marks of a professional is knowing that there is always more to learn.