For some reason, the last interview I did (on Squadcast, a first for me), left me with the guest’s track barely audible. I’ve had it in Descript and tried to mess around with it, but with no luck. I did turn on Studio Sound (at 100%).
Here’s my question: should I just bring the files out of Descript, put the whole thing in Audacity to normalize the audio track(s), then bring it back to Descript and essentially start over? Or do something else with it to get the volume up?
Thank you in advance for any suggestions or advice!
Hey, this is what I would do, with the separate guest track, making sure that there’s not a loud spike somewhere that throws the normalization off. Plus, you could just raise it manually to your liking after normalizing.
Just make sure you’re working with the highest quality file to begin with. Meaning, non-compressed, if you can get it.
This is what I’d try. Auphonic uses a process called Loudness Normalizaton which is slightly different to standard Normalization. But the risk with either normalization methods is that when you raise the levels by the amount you will need to then you risk increasing the background noise too.
Auphonic has a noise remover that can help counter this at the same time as it’s loudness normalizing (make sure the box is ticked to make sure this filter is on).
Audacity and Descript also have Loudness Normalization (when you are setting your audio to 19 or 16 LUFS that’s when you are loudness normalizing). Although if you use Audacity you’d need to use the noise reduction afterwards as a second step.
I’m not sure how Squadcast works but does it give you a separate video file? Could you drop that into Descript (or Audacity) and use the audio from that instead of the local audio files? Sure everything might be on one channel but if the sound is better than what you have now it could save a heap of work.
Lots of roads to explore. Thank you, @Jey - so nice to hear from you. I will see if Ihave the Squadcast video file saved, or if I just have audio files. Then dive in.
Forgot to mention one last free “one button solution” to try.
Adobe have a tool called “Enhance Speech”. It’s like Descript’s Studio sound. You might have to join the free Adobe Podcast Beta. Worth a try and I’ve had good results.
There might be a time limit on recordings (an hour I think) you can upload at one time. So you might have to split your recording - or try a short sample to see if it can help.
Great learning opportunities are also really frustrating.
If you are chasing a tight deadline - do what you know. eg Audacity
If you want to learn more about Descript for a future situation - dig in deeper and stick with Descript.
Some things you may already know about Descript and a few questions to clarify your problem
Descript auto levels on import.
Did you do an automatic internal switch from Squadcast to Descript for editing, or did you download and import the file(s)?
Did you end up with two audio files? You can check the media to see if there is a sequence (basically the same as multitrack in Audacity). Right-click on the timeline (you may have to reveal it (bottom left) and choose Edit Sequence.
If you have two tracks you are in good shape to improve things.
Studio sound does adjust levels some but also deals with background noise, reverb and the like - remember it has already been auto-levelled on import. You can apply Studio Sound to the whole composition or if you edit the sequence (use the Edit menu to select, or right-click on the timeline) you can apply Studio Sound and other effects and adjust levels to just that track. This includes auto level clip - which will deal with the whole track for you. Could be problem solved.
A bunch of pointers have already been shared in previous posts and mine - it just depends what works best for you.
Here is a video that will give visuals of some of the things I just described.
Let me know if you are still stuck, or confused by what I said.
Wow, friends! Your suggestions worked. Once I confirmed that indeed I had two tracks to work with, I was able to do it in Descript, upping the DB volume, per Steve’s suggestion. The volume is such that I can now edit it, and then, per Jey’s suggestion, send it right on over to Audacity and tweak the normalization when I get there. I did also check out Adobe, which I’d used before at Jey’s suggestion, and it is 30 minutes/500 MB max upload, one hour per day. So be it.
A lot of people are talking about Audacity, which is awesome, free and what you currently use. If you want a lot more control, you may want to move to DaVinci Resolve, which I find much more powerful, robust and dynamic. It’s much more akin to Adobe products, but costs zero dollars. Version 18 (They’re always coming out w new versions) has a great normalization to YouTube feature in the export panel. It has saved my butt MANY a day.
Raising the gain (The volume) adds noise and DaVinci has a great denoiser. There are a ton of tutorials on YouTube for ‘support and learning’ (For free) It’s fiddly, and complicated, but if you’re using Audacity, you’re halfway there.
And just for fun, DaVinci is much more stable than I’ve found Adobe and Audacity to be. I have a very powerful computer (I also edit video) and they both crash a lot when you start getting too fancy, and DaVinci is very stable. This is not a paid endorsement. LOL
Thanks, @MoeP - definitely sounds worth checking out. I will say, having paid for 2024, I am somewhat tethered to Descript. But that’s not all I use, not by a longshot.
