Possible to sound better?

Sup :]
Firstly Happy New Year :smiley: Hope everyone had a great holiday. Over the holiday I bought myself the lovely se electronics se2200a ii, and the scarlett 2i2 audio interface. I love my microphone, but I admit audio tweaking isnā€™t my strongest point. Itā€™s a constant struggle for me, and itā€™s killing my mood to record. I believe the problem is that I am somewhat a perfectionist, but lack any real understanding of how audio works. I am hoping to get any tips on how to improve, or is there any need to improve?

Below are my settings, and the order I do my EQ:

  1. Noise Removal: 30 (db) noise reduction - sensitvity 3.00 - Smoothing bands 0
  2. Bass Boost: 36 (top slider) : 34 (bottom slider)
  3. Treble Boost: same as above
  4. Compressor - Threshold -20db : noise floor -40db : ratio 6:5:1 : attack time 0:10seconds : release time 1.0 seconds : make up gain for 0 db (ticked) : compress on peaks (ticked)
  5. Normalize: -1db
  6. Limter: -3db
  7. Normalize: -1db

I do lets plays on YouTube, and enjoy my game audio to be loud (like a movie) but not overpowering my voice, and my voice not overpowering my game audio. I use auto duck, but try my best to make sure it doesnā€™t sound I use it (if that makes sense). Sometimes I feel that I nailed my audio, only to find out later that I donā€™t like it, or think its any good. I have been bouncing back and fore between making my voice -3db, or -1db.

Here is my clip that I am working on now: https://soundcloud.com/theecondgent/test

I have my audio interface gain on pretty high. If I lower it anymore the green light wonā€™t appear. My microphone is at full 100%. I know that sounds insane, but the audio coming out is SUPER low.

If you have any information on how I can adjust these settings, I would seriously appreciate it because I would really like to have good quality audio. Itā€™s also driving me somewhat insane, as I have been doing videos for almost 9months, and unable to get this down ><

Thanks for reading !!!

You canā€™t like your microphone that much with that list of corrections and 30dB noise reduction.

Post a sound sample before you do anything to it. Record, Stop, Export. Leave two seconds of silence (hold your breath) and announce normally for ten seconds.

Use WAV (Microsoft).

I need to listen when I get home.

Koz

We can address the low volume first.

The 0dB/-10dB switch should be at 0dB. -10dB is for instruments like micing a drum set, trumpet or a Boeing jet.

This is a side-fire microphone. You should be talking into the red sE logo on the side, not the end like a garden hose. Itā€™s a cardioid microphone, so speaking into the back just doesnā€™t work.

Since you donā€™t have overload problems (apparently), you can switch off the High Pass Filter. Thatā€™s the switch on the right. You do not want the bent line.

Yes, the green activity light should come on when you present, but the world will not end if it doesnā€™t. A much better indicator is the bouncing Audacity Recording Meters.

Thatā€™s ideal, although I recently produced a good recording lower than that. Set the upper (Recording) meters to Gradient/dB/Automatic. Record in the Yellow areas and start to get concerned in Orange. You should not ever run the meter all the way up and have it turn red. That will create unrecoverable crisp, cracking damage.

You should make sure all the Windows sound processing is turned completely off and the sound control panels are all the way up as is the Audacity volume controls. Do everything with the controls on the 2i2. Do Not use any Windows 20dB boosts or helper settings like that.

http://manual.audacityteam.org/o/man/faq_recording_troubleshooting.html#enhancements


So thatā€™s basic setup for normal recording. Audacity doesnā€™t apply filters or effects on recording, so whatever you do during the performance is ā€œnatural.ā€

Thatā€™s what you should post as your ten second Mono, WAV sample.

Also, locate a sample of performance not by you that you like. ā€œI hate the way it soundsā€ doesnā€™t tell us anything. Compared to what?

Thanks man for replying :slight_smile: Itā€™s not that I dislike my microphone. I just lack any real knowledge of how to make good audio. Never had professional equipment so itā€™s all very new to me.

I always try to talk into the sE logo when recording, but my desk is somewhat small. I have to angle the monitor and the microphone so I can achieve this while being able to see the screen. I made the adjustments on the microphone, and on Audacity itself.

I typed in the microphone name, along with commentary into YouTube and Google. Looked around for a bit, and found these good examples.

1: https://www.youtube.com/watch?v=XMlqpwNNWz8
2: https://www.youtube.com/watch?v=bpmIdWQBM0I
3: https://www.youtube.com/watch?v=kwPDQB0EVbY

If I could achieve that and having good balanced audio, that would be fantastic. My audio isnā€™t terrible right now, but from looking around I see it can be better.

Hereā€™s a sample without any effects at all: SoundCloud - Hear the worldā€™s sounds

Again thanks for your time :smiley: And sorry for the late reply!

Hereā€™s a sample without any effects at all:

Getting there. I hear moving around and thumping and clunking in the silent portion. Thatā€™s supposed to be just the noise the microphone and the room is making. We canā€™t use it for noise analysis with extra noises. Hereā€™s a mini-class I wrote.

http://www.kozco.com/tech/audacity/TestClip/Record_A_Clip.html

And yes, please post on the forum instead of SoundCloud this time. We have to capture the sound from SoundCloud and we only have to copy the files from the forum. ā€œTestingā€ works, but read something for real. Read the side of the milk carton or cereal box.

ā€œCorn Flakes helps build strong, young bodies with all the vitamins needed for a healthy, balanced ā€¦ā€

Koz

I like the first sample clip. I could listen to that for a long time.
MissBeccyMay needs lessons on how to run her microphone without breath noises and popping her P sounds. You can use a pop and blast filter to help. Thatā€™s the little black tennis racket thing between the actor/singer and the microphone.

So yes, I got the idea.

Itā€™s not unusual for people to want us to fix their acting with filters. We canā€™t do that. In the third example, the announcerā€™s voice is OK, but I couldnā€™t listen to that acting very much without wanting to throw something at the screen.

Waiting for that next clip.

Koz

Hereā€™s the clip :slight_smile: I read over everything you said again, and I made adjustments to the microphone position. I made the red sE logo go on the side, while I talked. I got the tennis racket thing setup, but I can add another pop filter over it, as I bought one.

Out of interest why didnā€™t you like the last one? Personally I thought all three were high quality. Just asking so I know on how to improve my audio by understanding any faults in the examples provided.

Thatā€™s it. I changed the volume slightly with Effect > Normalize and I did a very gentle noise removal. Attached: Last sentence. The corrected clip sounds almost exactly like the original, so there are no correction errors or sound damage.

So just what youā€™re doing will pass the ACX AudioBook technical publishing standards.

Thatā€™s the good news.

I couldnā€™t understand some of the words in the performance. You have an odd clip to your speech that makes some of the words hard to understand. That may be why you couldnā€™t make the presentation sound clear no matter what you did. Thatā€™s acting and there is no filter or effect for that.

This is how I corrected your clip. It was very close as it is and didnā€™t take much work at all. You have a clear, quiet recording system. You may want to Normalize to a higher volume (-1 ??) since youā€™re not publishing to an audiobook. Audiobooks have strict limits at -3.

Noise Reduction
ā€“ Drag-select Room Tone, silence or the flat area between spoken phrases.
ā€“ Effect > Noise Reduction: Profile
ā€“ Select the whole clip or show by clicking just above MUTE.
ā€“ Effect > Noise Reduction: Settings 6, 6, 6 > OK

Volume Correction
ā€“ Select the whole clip or show by clicking just above MUTE.
ā€“ Effect > Normalize: [X]Remove DC, [X]Normalize to -3.2 > OK

Everything else is mixing with the show sound. You should be able to change the volume between the show and your voice in the video editor. You shouldnā€™t need to keep re-exporting at a different volume.

Koz

Screen Shot 2016-01-01 at 23.15.40.png

Thanks man for all the help :slight_smile: I had no idea I was hard to understand. Itā€™s something that I will work on for sure. Quick question: Do I throw out all my steps in the first post, just using normalize and noise reduction? Should I add some bass and treble, or would that be too much?

Thanks again buddy.

Do I throw out all my steps in the first post

The first sign of trouble in your post was a very high noise reduction value. Itā€™s impossible to make even the new Noise Reduction tool sound gentle and clear with high numbers. It can give bad cellphone or talking in a wineglass sound. You put just enough Noise Reduction in to make the performance acceptably quiet and then stop.

As in my post, simple Noise Reduction and Normalization will produce a sound track that people will buy (given the voice is OK). So using that as the starting point, you can apply equalization filters to increase bass, crispness, etc.

The equalization and other tonal quality tools (crisp, muffled, boomy honky) are up to you, but fair warning how you use them depends on your speakers or headphones. I use powerful bookshelf speakers and a floor-mounted bass cabinet for listening and balancing. In general, if you can hold your speakers in one hand, they may not be the best for sound balancing.

If you know your audience is all using the same or similar speakers or headphones, then you can balance for that. Equalizing and finishing is not one single, simple goal. Also, you donā€™t want a sound system so specialized that it misses problems. You post a track that has rumbling, thumpy sounds behind your voice but you canā€™t hear them.

You can use Effect > Compressor. That can reduce the variations in volume during your performance. Note this is the AudioBook version. If you work louder, say at -1 instead of -3.2, use your number in Normalize. You voice will become denser and louder. You can make it denser yet by increasing the Ratio: value. UNDO the old one before you apply a different one. Multiple effects can give crazy sound.

Audio Compressor
ā€“ Select the whole clip or show by clicking just above MUTE.
ā€“ Effect > Normalize: [X]Remove DC, [X]Normalize to -3.2 > OK
ā€“ Effect > Compressor: Thresh -20, Floor -40, Ratio 2:1, Attack 0.2, Release 1.0, > OK
ā€“ Effect > Normalize: [X]Remove DC, [X]Normalize to -3.2 > OK

Koz

The posted examples are better than you think. Theyā€™re very different.

I liked the first one because past the good quality and clear voice, he seemed to be really excited about the game and he knew how it compared with other, similar games. He pulled me into the game with him. I have crystal clear memory of what he was talking about, and Iā€™m not a gamer. I have no trouble listening to him at all. Itā€™s hard to beat that combination.

The woman singing has a very nice voice and bad microphone technique. You should not be so close to the microphone that it picks up extra Puh and PP popping sounds, whether youā€™re singing or announcing.

http://kozco.com/tech/audacity/clips/PPopping.mp3

ā€œI really do aPPPreciate itā€¦ā€
ā€œReal Life is a PPPodcast aboutā€¦ā€

I didnā€™t tell Chase I was going to use his podcast as an example of bad announcing.

The last sample has an OK voice but terrible acting. He sounds like heā€™s trying to act superior and clearly has no idea what heā€™s doing. He can do that without me.

The local radio station has two woman announcers. One has a slightly edgy voice and tries too hard to be a ā€œRadio Announcer.ā€ She pulls me out of the story and I pay attention to her instead of the show. The other has a natural presentation, rhythm and smooth voice. I could listen to her read the phone book. Same station, same show, same microphones.

Koz

Thank you very much :slight_smile: I will sit down, and go through my settings to adjust. You have been of great help! Iā€™ll bookmark this page, so I can keep referring back to it.

Thanks again!

I did another test, and wanted to show you the results. Its amazing what microphone position, and a few tweaks can do :slight_smile: I did write down a thank you post, but it doesnā€™t seem like it went through.

So thank you very much for the help :slight_smile:

Newbie posts are moderated and inspected to make sure youā€™re not trying to sell us ā€œadult entertainmentā€ or new kitchen cabinets. We do stop after a while.
Koz

The overall sound quality and clarity is much better with those new settings. Thereā€™s just a hint of P popping and breath noises. You may get rid of that by backing away from the microphone just slightly.

I think your intelligibility improved too, possibly because youā€™re speaking a little slower. You should ask other people about that. Iā€™m much better with the technical adjustments than the acting.

Koz

I like to listen to Sarah Vowell audiobooks. She has a slightly odd presentation, too. But she just takes her time with pronunciation. After a while, you just get used to it.

http://kozco.com/tech/audacity/clips/SarahVowell.wav

Koz

Hey, just replying back to ask if you ever used the noise gate plugin for audacity? You helped me out leaps and bounds last time, and Iā€™ve even upgraded my audio set up from the last time we spoke (audio mixer with EQ). Everything is fantastic. While itā€™s not really a problem (but it does cost me a lot of time to edit) there has been breathing sounds, and other general noises in my audio. This has always been the case, but I simply just manually edit them out. I donā€™t noise gate them out, as I want to keep the highest quality possible in my audio.

If thereā€™s a way to noise gate out the breathing sounds without losing quality then that would be brilliant. I understand simply moving back from the microphone would help, but the audio doesnā€™t have thatā€¦radio quality, I suppose. I do have foam placed around my room, but I donā€™t think itā€™s enough because the quality does drop if I move myself further back.

If you could help me with the settings, or give me general information that would be amazing :]

I should also point out, I still use all your settings aside from the bass and treble boost because my audio mixer does that for me.

Noise gating is very tricky, post production. I do it pre production. To light and the noise comes in, to heavy and your words get chopped. You will need to do it different each time, with each recording. One size does not fit all in this situation. Since you can only apply it after each recording, it will take some playing around with, to get as close as you can. Make sure you render your sample to see if it fits your needs before you apply the effect to the whole file.

I see. Iā€™m not brilliant with sound stuff, so this might be tricky. Do you have any other advice to stop breath sounds, and mouth sounds? To be honest odd mouth sounds is the biggest killer for me. The only solution to back away from the microphone and buy more sound dampening foam?

Thanks for the reply!

As above, Noise Gate is not easy to use and if you use it wrong, it can give you chopped, robot voices, pumping sentences or tails on your words. ACX calls that Excessive Processing (one of a variety of things that can cause that) and if itā€™s bad enough, they will bounce a submission.

You are in fuzzy territory here. ACX objects strongly to heavily processed performances, but they also donā€™t like ā€œdistracting noises.ā€

Each uploaded file must be free of extraneous sounds such as plosives, mic pops, mouse clicks, excessive mouth noise, and outtakes.

Why? Extraneous sounds can be distracting to listeners, and outtakes sound unprofessional. Each can elicit bad reviews and bad reviews can hurt sales!
>

It seems to be between you and them. I have no horse in the race beyond passing basic recording standardsā€”usually hard enough.

As has been posted multiple times, if your basic performance isnā€™t ā€œclean,ā€ making clean out of what you have in post production is very difficult. The answer to a posted question, ā€œCan you clean this up?ā€ is almost always no.

~~

And just because I donā€™t think we ever actually hit this, are you reading for AudioBooks?

Koz