Carrying over to final normalization, compression, limiting

[River]

Hi, Dana, well, another bind has brought me back. ) I was mostly pleased with the audio capture of a story I just finished. The editing is all done, with some noise reduction, a transparent gate, and a little compression applied. I finalized by exporting as 24-bit WAVs and then normalizing and compressing the tracks till they were mostly in the ideal RMS range, peaking out around -3. (This is self-released so I'm okay not nailing it as long as I learn how to nail it next time.) In retrospect, I don't know what rendering and then reinserting in new project did as they were all mono voice tracks, anyway. But somehow after doing that I'm getting a lot of noise in my files, the type of thing I'd expect if I recorded at a too low signal. I was going to release it anyway, but after rendering to the final 16-bit WAVs, it's much worse with pronounced clicking artifacts.

I wondered if it perhaps was a too low signal after all that I didn't notice because I wasn't monitoring at a loud level. But when I go back to my edited, "pre-mastered" tracks and turn the volume way up, it sounds good. Any ideas how to address this?

Happy Thanksgiving, everyone.

[Dana]

I have to run out and buy a battery for my wife's car. I will address your issue when I return. One thought, was your computer fan running louder then normal? Sometime when you render and you have a lot of stress on the computer or ram, it will add noise as well as clicks into the recording. Also, did you add any dithering when you exported?

[River]

Interesting, I did have a lot running on the PC though I don't recall how it sounded. I rendered as selected media items rather than master mix, so dithering was not an option.

[Dana]

Interesting, I did have a lot running on the PC though I don't recall how it sounded. I rendered as selected media items rather than master mix, so dithering was not an option.

It is highly recommended that before you do any recording, or rendering, you re-boot your system and then close any programs that are running in the background, that are not needed. This will put all the ram back where it belongs. I just went back and read your post again. When you render stems or selected tracks, any plugins you are using on the Master Track, get bypassed if I am not mistaken. Rendering stems is mostly used for sending your finished tracks, to a Mastering Engineer. The video below is very educational concerning stems and Reaper. It is 10 mins long and may help you figure out what happened. If you want, you can send them to me as well and I will check them out later this afternoon or early am.
https://www.youtube.com/watch?time_continue=389&v=SAkAvvFj7WQ

[River]

Eureka. Rendering as stems was the culprit. Some marathons end in steep upward hills; Reaper ends with its rendering menu. It is not intuitive to me.

I re-arranged the files so that they were on separate tracks, but beginning where the previous track ends. Then I exported as a master mix for pre-mastering, then separate tracks in the mastering. I could've done it as project regions too (I did that for the "mastering"). That way it exported preserving the clean-up effects. Overall I'm happy with it, considering my limitations and self-imposed deadline. I ref'ed it with another audiobook. The "professionally" mastered one sounded warmer but the difference was not jarring. Mouth noise haunts my performance but it is not ruinous. I'll send you the link.

If I understood the statistics, most of the audio is actually a little lower than ACX RMS standards. I also don't have a limiter with an obvious tool for setting the maximum output. Any tips you have on how to get purposefully spot-on would be helpful.

[Dana]

Eureka. Rendering as stems was the culprit. Some marathons end in steep upward hills; Reaper ends with its rendering menu. It is not intuitive to me.

Just so you know, I do not like to have to look up big words, nor do I want to drive to WV, to hit you over the head with a dictionary. (However I do miss WV.). I am glad you were able to figure it out.

I re-arranged the files so that they were on separate tracks, but beginning where the previous track ends. Then I exported as a master mix for pre-mastering, then separate tracks in the mastering.

I may be misunderstanding your statement, but it seems like you are doing the same thing twice. If that is the case, then you will get a lower quality of audio, due to over processing.

I could've done it as project regions too (I did that for the "mastering"). That way it exported preserving the clean-up effects. Overall I'm happy with it, considering my limitations and self-imposed deadline. I ref'ed it with another audiobook. The "professionally" mastered one sounded warmer but the difference was not jarring. Mouth noise haunts my performance but it is not ruinous. I'll send you the link.

I look forward to the link. Stop, take a breath and remember that deadlines have no place in recording audio. That is why Artists, spend months in recording studios. It has nothing to do with experience, it mostly deals with someone feeling being rushed, and their true talent suffers, as they can not preform, in their natural way. I have listened to your audio on $60,000.00 speakers. You are much better then you think you are!

If I understood the statistics, most of the audio is actually a little lower than ACX RMS standards. I also don't have a limiter with an obvious tool for setting the maximum output. Any tips you have on how to get purposefully spot-on would be helpful.

There are many third party plugins, concerning limiters. I have three that I paid over $1,200.00 for. Honestly, the ones that come stock with Reaper, do just as well. Look for the "Volume Adjustment", "Peak Limiter" and "Master Limiter" plugins that come stock with Reaper. They are simple to use and I can assure you, 99.9% of Mastering Engineers, would not be able to tell the difference between the Reaper plugins, compared to the IK mastering Suite bundle. I get hate mail all the time, contradicting my statements, but, it is what it is.

Put ten Mastering Engineers, Chefs, Carpenters, Doctors, Day Care Helpers in the same room, and you will get 10 different opinions, every, single, time. If you want to tame the peaks, use the peak limiter in Reaper. After all is said and done, when you are 100% finished and you want to render the project, slap the Master Limiter on the Master Track and let it eat.

Just so you know, you do not need to have a problem, in order to post here. You are more experience then you know, and I think, you have a lot of advice you could give to others. Try and think back to the first time, someone placed a mic in front of you. That is what we are all about here. Taking someone with no experience, to the point of recording professional audio, regardless if it is just speech, or laying down 30 tracks. Happy Turkey day. For the next 12 hours, I will be doing what this guy is doing, in the pic below. After that, I will be cooking.

[Dana]

OK, I have some time as the dinner is taken care of and the smoke alarms are silent. I tried to recreate your situation and this is what I found. It had my one of my CPUs slammed @ 100%. It was all due to the plugins. Another thing to keep in mind, that most songs are 2.5 to 3.5 mins long. Your story chimes in around 70 or 80mins, I forget.

The best way to do this is to record your intro, save the project as intro. Record the credits, save the project as credits. Record your chapters one at a time and save them as chapter 1 through ? Make sure that as you are recording, if you make a mistake, try to re-read that part in the same setting. If you do not have time to read the whole chapter at one time, then wait till you can. So once you have all your recordings done, remember they are still in the "project form", open one file at a time. Put the audio statistics plugin on the master track and bring it up.

Go to render on the file tab and set your export settings. 44.1, 196, mp3, mono. Render that file. DO NOT HIT CLOSE when it is through. All you want to do is see what RMS level it is at. Once you know that, hit the BACK button and place the Volume Adjustment plugin, as the first plugin on the master track, with the audio statistic plugin underneath it. I try and shoot for a -20RMS level, this will give me 2dbs of head room on the loud side and 3dbs on the low side.

Open the volume adjustment plugin and add what you need to bring it up to 20rms. you do not need to use the + key. Now where it says max level, put -3.5, here you must use the - key. Now bring up the audio statistic plugin and render the file again. The only thing you are looking for is the rms level, and the max peak. Again, hit the back button and not the close button. If everything look good, place the reaper gate in-between the volume adjust and the audio statistic plugins.

Leave everything as default except for the threshold adjustment. Set this to a -40dbs. Bring up the audio stat plugin again and render the file. Now you can check your noise floor, rms level and max peak. If it all looks good, uncheck the box for the audio stat plugin and render the file as the name you want attached to it.

If you follow this process for each file, you will never have a problem. I will try and do a video for you if needed. The pics below show my cpu usage with plugins on, as well as off, for all 20 tracks. Regardless of what plugins you use, they are all cpu hogs. That is why you see a lot of videos that will have a compressor master track. Anything that needs compression, is sent to that track first, and then to the master track. Click the pic to enlarge them.

[River]

Wonderful. When I have time I would like to give a shot at refinishing the files with the process you've described.

Just so you know, you do not need to have a problem, in order to post here. You are more experience then you know, and I think, you have a lot of advice you could give to others.

I appreciate that...I honestly haven't seen anything else I could speak to, but, when I bring up an issue, I try to be specific both in explaining it and how it gets resolved, so that onlookers with the same question can benefit.

I had considerable help on the music project from a sound engineer friend. I have learned a lot, but his involvement was crucial for EQing.

Stop, take a breath and remember that deadlines have no place in recording audio.

I have found the only way I finish anything is with deadlines! And then I've learned my lessons much deeper when the mistakes get out in the open. The key, however, is using realistic deadlines.

[Dana]

I have found the only way I finish anything is with deadlines! And then I've learned my lessons much deeper when the mistakes get out in the open. The key, however, is using realistic deadlines.

Then you have achieved what the other 98% could not. Welcome to the 2% club.

[River]

I used the Volume Adjustment & statistics with success. What is the best field in Audio Statistics to gauge noise floor? I am not sure it is below the -60 required by ACX. The volume adjustment unsurprisingly made the white noise louder.

The vocal sounds pretty warm too, as in over-compressed, not in a really flawed way, but I think that was probably how I was micing with the SM57 and pop filter.