Improving badly recorded audio can it be done?

[Eugen Brinzoiu]

Originally Posted by Noa Put

The problem with weddings is that once a ceremony is in progress you can't just interrupt the vows if your hear your audio system fails, the only difference between being able to monitor the sound and not being able is that you can hear your audio fail but you can't do anything about it, so much for being a "professional" :)

Noa......you're absolutely right!

[Ana Conlon]

Originally Posted by Rick Reineke

"if I hired someone could they do much more to improve it?"
- Probably, depends how bad it is. It still would not be anywhere near as decent captured sound. Post a few medium resolution MP3 clips, there's plenty of Dropbox type file sharing sites around. Some of the members here and over at the DVX user audio group have iZotope RX Advanced, which can do some pretty amazing stuff.

Hi Rick I am going to try and see if anyone at DVX can do anything to improve it but I fear there is very little that can be done! I appreciate the recommendation. If I do get help here I will post the improvement for others to reference.

[Ana Conlon]

Originally Posted by Rob Cantwell

Yeah echo is pretty hard to resolve, also cutting out the audio between the responses only serve to amplify the problem, I've had some small success with Audition capturing the noise print sample and using that to reduce it somewhat, but it seems you've already done that. Cant see much to be done after that, as it's baked in.
I'd leave it all in noise and all rather than try to isolate the responses, it sort of brings attention to the audio.

good luck with it

I totally take your point about the amplify sections. I just worked on it for ages and I can't even hear it anymore. I will readjust the amplified sections. Thanks.

[Ana Conlon]

Originally Posted by Andrew Smith

I don't think this one can be greatly improved at all.

The in-camera mics on your DSLR simply aren't all that flash to begin with and you don't have the necessary proximity to the source that you need to keep when removing the room reverb / echo. (There has to be a threshold between the audio to be kept and the audio to be removed.)

Worse, when working on the file it sounds like I am uncovering the audio "mud" that you get from .mp3 audio compression (where the bits you don't normally 'hear' are discarded). This in addition to what I suspect is an auto-level function on the camera mic that is also playing havoc.

I think everything is working against you when it comes to this audio sample.

I've had a quick try with the latest tech "dialogue isolation" which separates based on voice / non-voice audio. Better but not perfect. Zipped file is attached. (Then there is the accent these people must be speaking with. :-P LOL) See what you think.

Andrew

Originally Posted by Gordon Hoffman

I played around a bit with this but I don't really know what I'm doing. I just like pushing buttons.

Gordon

Thanks for your attempts with this. The overall effect is very similar to the background remover that FCPX has build in....I am looking at echo remover from Fxfaxtory at the moment to see if this will give any further improvement.

[Bernie Beaudry]

If you can upload the file before any attempts at noise reduction (So the original file) I'll take a go and see what I can do. . I have RX6 and Acon Deverb and I'm pretty practiced on both. In the interim I'll see what I can do with the version you uploaded before.

[Bernie Beaudry]

Just did a quick listen on headphones and what you accomplished so far isn't too bad. I'll still take an attempt on a original file just to see but I think you're on the right track. You just need to work on things being more cohesive so you don't notice the processing so much. You don't want the background to fluctuate so much that it calls attention to itself.

[Ana Conlon]

Originally Posted by Bernie Beaudry

If you can upload the file before any attempts at noise reduction (So the original file) I'll take a go and see what I can do. . I have RX6 and Acon Deverb and I'm pretty practiced on both. In the interim I'll see what I can do with the version you uploaded before.

Thanks Bernie you're a star!

Today I've messed around with echo reducers but haven't had any luck with that at all.

I have audio from 3 media sources, there was a zoom on a podium close by and then I had two DSLR cameras. One I was moving about so there are some pops and clicks...and the other was stationary.

I'll post all 3 here naming them accordingly.

https://drive.google.com/open?id=0B1...XdMSzlpd1JFV1k

P.S. After pulling all these files I'm starting to think they sounded better than when I started messing around with them! I only have final cut pro X for audio editing and audio mastering wouldn’t be my strong point!

[Bernie Beaudry]

I'll take a look at these later today. I like a good challenge! It helps me learn the software and figure out what's possible. I'm making no promises as to the final result but I'll do my best.
BB

[Ana Conlon]

Thanks BB

but please don't put too much time into it...I would hate to eat up your day! I have worked on it again today myself (taking some tips from here) and its probably as good as its going to get!

[Bernie Beaudry]

I'm making good progress. I think you'll like the result when I'm done. Don't worry about my time this is a great piece to train on. I'll upload them when I'm finished.
BB

[Bernie Beaudry]

Originally Posted by Ana Conlon

Thanks Bernie you're a star!

Today I've messed around with echo reducers but haven't had any luck with that at all.

I have audio from 3 media sources, there was a zoom on a podium close by and then I had two DSLR cameras. One I was moving about so there are some pops and clicks...and the other was stationary.

I'll post all 3 here naming them accordingly.

https://drive.google.com/open?id=0B1...XdMSzlpd1JFV1k

P.S. After pulling all these files I'm starting to think they sounded better than when I started messing around with them! I only have final cut pro X for audio editing and audio mastering wouldn’t be my strong point!

Sometimes less is more. What I'm finding is the low volume of the vows doesn't bother me so much or the echo. Its a church after all. Its the random baby crying, the thunks from moving stuff, etc. that are closer to the mic than what you want to capture. I can work on those some and between the three sources I think you'll be able to put together a smoother sounding piece that will keep the viewers in the moment.

[Bernie Beaudry]

Originally Posted by Bernie Beaudry

I'm making good progress. I think you'll like the result when I'm done. Don't worry about my time this is a great piece to train on. I'll upload them separately when I'm finished.
BB

Well I'm going to call it for now. Let me know how these cut together for you now. It was getting very time consuming trying to take out all the random sounds. I would checker board between the three sources and try to keep the edits cross faded so the background stays at a consistent level.
BB
https://drive.google.com/drive/folde...0k?usp=sharing

[Ana Conlon]

Thanks so much this must have taken ages! I will be able to switch between the audio here depending on what's clearest for each section. This is my best chance of achieving a good sound.

I'll post an update on how I get on! I really appreciate your help and indeed everyone who has posted on this thread! Thanks a million!

[Bernie Beaudry]

You're welcome! It was great training and I learned a lot! Most of the time was spent figuring out what worked better and how far to take it. If you edit primarily with speakers I suggest you listen on good headphones as well to really hear artifacts or other problems. I'm looking forward to hearing how it turns out!
Best!
Bernie

[David Peterson]

I'd put a lav on the celebrant, and bride as well. Not just the groom.

Best man too if you can! And bridesmaid.

That way if the groom's fails in the future, you always have the celebrant or bride's one to use as a nearby source to cover you for the groom.