Youtube videos commentary sound too quiet.

SinSpreader88 wrote on 2/23/2019, 12:03 AM

Ok so for reference I have my headphone volume on 50% and any youtube video I watch this volume level is more then competent.

When I edit a video with my audio it is vastly more quiet and I have to crank the volume on the Track to 5 and my headphone volume to around 76%

Then rendered my video sounds really quiet on 70% compared to the typical youtube video.

When I'm editing I'm trying to keep my voice from peaking in the rendered video. So I keep the volume in the green on the master bus.

What my actual question is though, is what exactly can I do to make my voice sound louder without blowing out the decibels and peaking the audio?

Am I missing something with how the audio works?

For anyone asking I'm using a mic with the levels at 65% If that at all helps you.

Comments

zdogg wrote on 2/23/2019, 3:53 AM

Hello,

Try experimenting with the included compressors and limiters, and don't be afraid to push it a bit more, Vegas levels just seem to be lower that some other programs when it looks like your going over 0 db, my guess is you're not, but that is not scientific, but the only way is to go ahead and do some tests, use short clips, like 10 seconds, so you can effect a good amount of varied tests without gobs of time on renders etc.

Modern sound is too often way over-compressed, and it's really too bad, sounds awful and no dynamics, plus, if your on the web (or even TV) the commercials will blow your speakers -- they are so loud compared to the rest of the audio material.

zdogg wrote on 2/23/2019, 3:53 AM

Any links?

rraud wrote on 2/23/2019, 10:18 AM

For YT with projects that are mostly dialog, I set finial program levels from -16 to -18 LUFS (integrated), with the max true peak sample value not exceeding 2.0 dBFS. Of course getting this takes mixing skills and knowledge of comp/limiters and such.

Auphonic can adjust broadcast levels automatically (ATSC -24 LUFS integrated), I recall trying it and it worked.. sort of .. some pumping and other artifacts were audible though, but that depends on the mix (garbage in, garbage out). The Levelator is another option to achieve more constant levels.

SinSpreader88 wrote on 2/23/2019, 1:18 PM

Any links?

< My video

> Video I'm comparing it too.

I'm thinking the issue is in my preamp. That the Gain on my preamp is lower then what I think my Mic levels should be at.

SinSpreader88 wrote on 2/23/2019, 1:19 PM

For YT with projects that are mostly dialog, I set finial program levels from -16 to -18 LUFS (integrated), with the max true peak sample value not exceeding 2.0 dBFS. Of course getting this takes mixing skills and knowledge of comp/limiters and such.

Auphonic can adjust broadcast levels automatically (ATSC -24 LUFS integrated), I recall trying it and it worked.. sort of .. some pumping and other artifacts were audible though, but that depends on the mix (garbage in, garbage out). The Levelator is another option to achieve more constant levels.

I'm guessing to accomplish this you have a pretty heafty preamp mixer?

fr0sty wrote on 2/23/2019, 8:01 PM

1. Make sure your input levels are as high as they can be without peaking when recording.

2. Use compression. A compressor plugin in Vegas will boost volume when it goes below a threshold your specify, or make it quieter if it goes above a threshold you specify. Pull up some youtube videos on compressors and google good compressor settings to use on spoken word vocals to get you started.

3. If you have background music, you can use a sidechain compressor. These take in an input source of audio and use that as a signal to know when to compress the audio. So, you can sidechain the music playing in the background, tell it to "listen" to the voice track, and this means when there is no voice, there is no compression. When there is voice, the compressor ducks the volume on the music a little bit. You can adjust attack and release thresholds to make that ducking sound more natural. I don't think Vegas comes with a sidechain compressor (Magix? Let's see that in 17 please!), but I do think you can add one, as Vegas supports VST plugins.

Edit: Here's how to sidechain in Vegas.

SinSpreader88 wrote on 2/23/2019, 10:09 PM

1. Make sure your input levels are as high as they can be without peaking when recording.

2. Use compression. A compressor plugin in Vegas will boost volume when it goes below a threshold your specify, or make it quieter if it goes above a threshold you specify. Pull up some youtube videos on compressors and google good compressor settings to use on spoken word vocals to get you started.

3. If you have background music, you can use a sidechain compressor. These take in an input source of audio and use that as a signal to know when to compress the audio. So, you can sidechain the music playing in the background, tell it to "listen" to the voice track, and this means when there is no voice, there is no compression. When there is voice, the compressor ducks the volume on the music a little bit. You can adjust attack and release thresholds to make that ducking sound more natural. I don't think Vegas comes with a sidechain compressor (Magix? Let's see that in 17 please!), but I do think you can add one, as Vegas supports VST plugins.

Edit: Here's how to sidechain in Vegas.

I have a Mic that is built to take louder noises so peaking really isn't an issue. I just want the overall audio to sound less dull.

ryclark wrote on 2/24/2019, 5:20 AM

Then definitely add some Compression. 😉 Also working closer to the mic would likely give a fuller sound.

rraud wrote on 2/24/2019, 10:36 AM

I just want the overall audio to sound less dull.

"less dull".. as in lacking high frequency clarity?
If that's the case, aside from general EQ, there are HF exciters available (free and otherwise), including a version from Waves modeled from the original "Aphex Aural Exciter" that many of us 'old school' recording engineers used in the 70s. The iZotope Mastering Suit 2, that was included with Sound Forge Pro 10, had a decent multi-band exciter. I also like rgc's High Frequency Stimulator as well, The rgb HF Stimulator (DX plug-in) is 'no longer available' but can be found with a little due diligence. There is also the X-cita by Elogoxa, which is modeled after BBE's Sonic Maximizer, another hardware classic.

Here are links to the Levelator and Auphonic, As I recall, the Levelator is a stand-alone application and the Auphonic process is on-line.
Levelator: conversationsnetwork.org/levelator
Auphonic: auphonic.com/he

Vegas never did support side-chaining.. requests fell on deaf ears since the audio only version. There are a few work-arounds, but none are very good. I prefer to apply volume envelopes to the limited cob job side-chaining.