I did an interview a while back on my cell phone. I don’t remember being like this bad because I was able to transcribe it into text with no problem. But I just put it into Audacity and you can hear me fine, but the guy I’m interviewing is an older guy and he just sounds like he’s mumbling. I mean you can hear what he’s saying but it’s not easy.
The guy I’m interviewing is kind of famous and I’m not going to be able to interview him again. So I’d like to make this work.
-Did Noise Reduction step (the right way)
-Did Compression then Amplify the volume back up
What app did you use? Have you done anything like this before?
That is all I have done. I am rookie.
Do you still have the original sound file—before you did anything to it?
I’ve done some good work with Voice Memos and Music Memos, but it was just me. I wasn’t doing an interview. Also, getting the work from the phone into my Mac was pretty painful. How did you do it?
I know just the recorder app that comes on the phone. And I’m sure I could find out exactly.
Yeah, I still have the original file on one of my old phones we’re actually on external hard drive before it went into Audacity.
It’s actually been off the phone and saved on to an external hard drive. Always did that in case I lost my phone or something that I would still have the interview. Then from there I just pull up the file on my computer and just dragged over and dropped it into audacity.
Post a ten second sound file. Five seconds of you in good quality followed by the damaged sound. Hopefully, you asking a question and then them answering. Take this from the original sound file.
Prepare the sound file. From a forum window, scroll down > Attachments > Add Files.
I know just the recorder app that comes on the phone.
You can get into trouble with some apps that try to “help you” by automatically suppressing what it thinks is Background Noise. That’s the first thing that came to mind when you described the recording. The guest is low volume and muffled but your voice is OK.
my cell phone.
It’s not an iPhone, is it?
If you’ve never done anything like this before, you might find your phone is directional—and it was aimed toward you. Most “Phone” microphones are on the bottom of the phone. Some are aimed down and they hope to goodness your voice is loud enough to be captured when you have it against your face. Some are actually aimed sideways toward where your lips will probably be. Both have advantages.
Okay I think it is attached. This is WAV out of Audacity but I did no editing yet.
Your comment about not using Noise Reduction could work because there is not background noise, so maybe it is lowering him as background noise. so maybe just use compressor to try to even our voices out then Amplify?
Set the threshold at his average db since he is quieter.
noise floor in the middle -50db
Ratio ummm maybe 8? I don’t know because I am louder than him?
Attack Time and Release Time all the way off to the left
uncheck the boxes
Then Amplify?
Anything else?
Not an Iphone lol a really old phone. I think LG G6 it was awhile ago.
I would have to figure out Level Speech 2 I don’t know anything about it.
I think that’s just his voice. He may be a smoker with a slightly husky voice, but his sibilants seem to sound OK. You can hear the phone/recording software struggling to keep his voice level and even. His background hash level comes and goes.
Attached some first-pass processing.
I think I can do a little better and there’s one tool I can’t find. The down side is you need to apply a complex pile of corrections only when he talks. Bring extra coffee.
He sounds like he’s standing next to a dot-matrix printer. Zeeee. Zeeeee. Zeeeee.
Most of the room-reverb, (which is reducing intelligibility), is in the bass range, (<222Hz).
So, IMO, cut back as much bass as you dare before compressing …
So are you saying right now the bass is around 222 Hz and needs to be lowered OR 222Hz is where I want to put it at?
Okay so under effects I see bass and treble.
It is at 0db is it better to go positive number or negative db?
Do I do it to the full file or just when he talks?
IMO apply the bass-cut equalization to the entire file,
then apply a compressor, (to the entire file).
You could manually separate-out the interviewee responses onto another track and process them separately ,
but that would be time-consuming, and where they join your speech would probably be conspicuous.
I have no idea what I’m looking at or what to do. I go under effects to base and trouble and it only is in DB’s. I go to the graphic EQ like your thing said and I just see a bunch of bars. I don’t see where the base is. I don’t see where the 222hz would be. I tried to attach a picture and it said the file’s too big. It’s just one picture of my computer screen.
I guess Walmart already writing this. I can ask about the compressor. How I usually do compressor is. I look for the quietest voice and I get the average of the quietest voice And I put the threshold at that. I leave the noise floor in the middle. If there’s a big separation between the volumes of the two people all put the ratio closer to the right. I leave a tact time and release time all the way to the left and I don’t click on the two things on the bottom.
