Jump to content
Sign in to follow this  
ErikH2000

Best amplitude for podcasting?

Recommended Posts

I can normalize my amplitude--no problem. I'm just wondering which peak is best, e.g. -1db? Obviously, I don't want it to clip and lose sound information. But I'm also worried about making my podcast louder than other podcasts. Because the listener could get irritated at me if my podcast is significantly louder than the previous.

 

The whole program is pretty even with amplitude because it is almost all conversational talking. Nobody yelling, no big sound effects, etc.

 

It seems like it must be a common question answered many times before, but I didn't have the right Google Fu to keep unrelated results popping up.

Share this post


Link to post

Your best bet is to just compress the whole thing at about a 3:1 ratio with the threshold set at around -20, adjust output gain as needed, and then just throw a limiter on there with a ceiling of -1dB and adjust that threshold to your liking. Let the listeners speakers do the work, don't worry about making it too loud or too quiet, thats what their volume knob is for.

  • Like 1

Share this post


Link to post

Here at Earwolf we really max it out. We compress it with a high ratio and fast attack, getting ~20db of gain reduction. Then we limit it with an output ceiling of -0.5 and an average peak reduction of maybe 2-3db. For dialog, you really want every word to be heard, there's not much need for a lot of dynamics. When I'm driving in the car listening to a podcast and they start mumbling or whispering and I can't hear it, it's frustrating.

  • Like 4

Share this post


Link to post

Thanks to Boognish and Brett Morris. You said a few things I didn't understand, but I think I will go to my software (Audition) and see if I can apply it. I've also learned in the last month with lots of editing hours, that my audio amplitude is actually up and down quite a bit more than I thought. E.g. an actor leans back from the mic for a moment to read his lines better and gets quiet as a result. And I hand-adjusted hours of audio to get the low amplitudes up with the rest. So I guess I should figure out my process better so that the software will do this for me.

 

I appreciate the answers, and another nice thing is that this post is now the #1 answer on Google for "best amplitude for podcasting", so other people that happen to phrase the question same as me will find your answers.

  • Like 1

Share this post


Link to post

Thanks to Boognish and Brett Morris. You said a few things I didn't understand, but I think I will go to my software (Audition) and see if I can apply it. I've also learned in the last month with lots of editing hours, that my audio amplitude is actually up and down quite a bit more than I thought. E.g. an actor leans back from the mic for a moment to read his lines better and gets quiet as a result. And I hand-adjusted hours of audio to get the low amplitudes up with the rest. So I guess I should figure out my process better so that the software will do this for me.

oh god yeah don't do that. that's what compression is for! that's why we set it so aggressively too. like is said, it's 99% dialog. it's not like music where it requires really really careful compression to bring out just the right overtones in the snare drum. it's just about making the voice smooth and loud, so i suggest hitting the compression hard. set a fairly high ratio, fast attack, and mid release, then adjust the threshold until you're getting a lot of gain reduction.. after that i also recommend a de-esser and an EQ to clamp down the sibilance and bring up the fundamental frequencies of the voices (4-5khz) and run a high pass filter (around 40-60hz maybe), and notch out any shitty sounding frequencies if you want. then run the limiter. if you do it right, you should very rarely have to manually adjust the volume.

 

and if the performer keeps backing away from the mic, tell them to stop! unless it's adam pally (whole other story)

Share this post


Link to post

 

and if the performer keeps backing away from the mic, tell them to stop! unless it's adam pally (whole other story)

 

This makes me think of Katie over at Nerdist. She has to deal with Pete Holmes. He's a funny dude, but you know setting EQs on that guy has got to be its own job.

 

P.S. Does Pally give you guys a hard time about his mic distance?

Share this post


Link to post

that reminds me, i owe pete some mustard.

 

no, pally kept screaming directly into the mic without backing off, just full blown peaking the meters. i slipped him a little note politely suggesting he back off if he's gonna yell, and he took one look at it and then passed the note to seth morris as if it was directed to him. then seth morris wrote back on the note "don't tell me what to do" and holds it up with this smirk on his face hahahah then the next minute adam is screaming even louder into the mic and staring right at me. point is i work with comedians

  • Like 1

Share this post


Link to post

Comedians love to fuck with the sound guy. I cringe whenever I see a comic drop the mic violently because i know the FOH guy is scrambling to kill that feedback and gently kissing his preamps back to life.

 

 

Also a question for Brett, I never put this together for whatever reason but, is your band's name a Doors reference? Cuz that's killer

Share this post


Link to post

Comedians love to fuck with the sound guy. I cringe whenever I see a comic drop the mic violently because i know the FOH guy is scrambling to kill that feedback and gently kissing his preamps back to life.

 

 

Also a question for Brett, I never put this together for whatever reason but, is your band's name a Doors reference? Cuz that's killer

yeah it is. props for noticing

Share this post


Link to post

Okay, I am coming back to this, because after rereading the above posts and trying to do it in Audition, there is still a lot of ambiguity to me. I may just bore everyone to death with my baby steps and not get any replies to this, but AT THE LEAST, I will have composed my thoughts by writing them down.

 

Here's the steps I follow. All thoughtful criticism welcome.

1. My raw audio has peaks that are mostly about -18db. I know that is too quiet and the necessary amplification means I'll have more ambient noise. The preamp is maxed. I have a Shure SM58 plugged into the preamp. The mic is inside of a porta-booth (foam-lined box) to shield it from horrible, ghastly room echoes. Having the mic in the box is what makes it so quiet. But it's impossible to read lines with your head in the box, so actors have their heads back a bit. This is temporary until I finish treating a smaller room that I can move into.

2. There are some spikes in there that are higher volume. I run a hard limiter at -15db to get the sample so step 3 will do something.

3. Then I normalize to 100% (0db). I figure I need step 2 and 3 for step 4 to work.

4. I run a single band compressor (in Audition it is the "Voice Leveler" preset). It has threshold -10, ratio 12:1, attack 0, release 100ms, output gain 0. This makes my average peak around -9db.

5. Then I normalize it again to 100% (0db).

6. Then I capture a noise print and run noise removal.

 

I understand some of what I'm doing, but other parts... hmm, maybe I'm screwing up, don't know. Anything above really making you cringe?

Share this post


Link to post

I am finally watching a previously live broadcast of CBB and noticed that Hogg Soggermann touches and moves his mic non stop in studio. Not even a single noise or squeak from the mounts. Nice setup.

 

If you want I can ask him to stop moving the mic during the recording for you.

Share this post


Link to post

Okay, I am coming back to this, because after rereading the above posts and trying to do it in Audition, there is still a lot of ambiguity to me. I may just bore everyone to death with my baby steps and not get any replies to this, but AT THE LEAST, I will have composed my thoughts by writing them down.

 

Here's the steps I follow. All thoughtful criticism welcome.

1. My raw audio has peaks that are mostly about -18db. I know that is too quiet and the necessary amplification means I'll have more ambient noise. The preamp is maxed. I have a Shure SM58 plugged into the preamp. The mic is inside of a porta-booth (foam-lined box) to shield it from horrible, ghastly room echoes. Having the mic in the box is what makes it so quiet. But it's impossible to read lines with your head in the box, so actors have their heads back a bit. This is temporary until I finish treating a smaller room that I can move into.

2. There are some spikes in there that are higher volume. I run a hard limiter at -15db to get the sample so step 3 will do something.

3. Then I normalize to 100% (0db). I figure I need step 2 and 3 for step 4 to work.

4. I run a single band compressor (in Audition it is the "Voice Leveler" preset). It has threshold -10, ratio 12:1, attack 0, release 100ms, output gain 0. This makes my average peak around -9db.

5. Then I normalize it again to 100% (0db).

6. Then I capture a noise print and run noise removal.

 

I understand some of what I'm doing, but other parts... hmm, maybe I'm screwing up, don't know. Anything above really making you cringe?

yes the normalization is a bit cringe worthy. if anything i would do the normalization first. normalize at -3db to be safe. always good to leave headroom until the very last step. this will raise your overall volume without doing any compression or peak reduction. now here is where you want to do the noise removal in my opinion, because your noise floor is at the lowest. if you wait until after compression and limiting, your noise is going to be much much closer in volume to the dialog, which makes it much harder to remove transparently. don't overdo noise removal, but you can be more aggressive at this early stage without affecting the dialog at all (make sure to set it right and look at the signals), and you might like the result.

 

now you can throw the compressor on. you can be aggressive, i maybe would set your attack to be a little bit slower than 0 but not much. adjust your threshold to your liking, but make sure it's low enough that you're really compressing everything but maybe the extremely quiet parts.

 

at this stage (or before the compressor) consider adding DeEssing and EQ. A good DeEssing can do a lot to smooth out the sound, especially with shitty mics. If you're not doing any EQ at the recording stage, you really ought to here. Cut cheap sounding and nasally frequencies, run a high pass to cut any unnecessary low frequencies. i like boosting around 5-6k, generally a nice bright spot for the voice.

 

now run your limiter as the last step. don't output at 0db though, do -0.5 or -1db to be safe. smooth out any peaks, experiment with how aggressive you run it and how much you like the sound. with just people talking, i think it's ok to brick wall it fairly hard. harder than music, at least.

  • Like 2

Share this post


Link to post

Brett, thanks for the advice. I understood most of what you said above, and changed my process a bit based on it.

 

Since last, I padded a room and put my mics in there, enabling me to get rid of the porta-booth box and get closer to the mic. So my starting levels are much better now.

 

I know I can get what I have to sound better, but I think I'm in the "good enough" zone for myself and most listeners.

 

-Erik

Share this post


Link to post

This is something I was wondering. I was listening to CBB and the loudness is absurd compared to most other podcasts. My main issue was that I have been compressing like you do with music. Your points are great and I am going to try them out on my next cast and see how it works.

 

Y'all use PT in the studio correct? Do you use the PT compressors, do you use third party plugins like waves, or, do you compress with outboard equipment?

  • Like 1

Share this post


Link to post

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×