I am editing some shows for the radio station that I work for. Mastering, if you will. I took a sample shot and you can see that when the DJ is talking the volume is considerably lower than the songs.
Normalization only raises the songs not the dialogue. These files are 4 hours long. Is there a way to bring up the dialogue without doing each one individually?
This is an ongoing project and I have to do at least one a week.
So you get the whole show in one single file? And your image here shows this original file, without any additional processing done on your part?
Could you upload a short sample so we can play with it?
Play with compresion and hard limiting would be my first suggestion.
Open the preview window and start to configure. I would suggest doing this in the effects rack so you can see your changes happening before actually destructively editing the waveform.
The simplest way is to set a single bander compressor. Work out the threshold by running amplitute statistics over the quiter vocals and get the average level. Then set that as your threshold and work the ratio until you see the vocal and music come closer.
The use hard limimting and set the maximum amplitude at the level you’re currently normalising to. Then work on the input boost until you see the quieter bits boosted sufficiently that they’re at the right level. I would suggest you do this in the effect panel so at least you can see what will be happening with your changes before you apply.
Be aware that in boosting the compressing then boosting the vocals you run the risk of allowing noise that was recorded into the voice recording session to be greatly boosted if there is “studio noise”
Second option is the mastering tool but use it carefully. Loudness maximiser is what you want. Again with the preview window open do it in the effects rack so you can see before you apply. Move the loudness maximiser up until the preview waveform looks more like you want it to.
Be careful with the mastering effect though, it’s a powerful tool that if used correctly can make good audio sound great but if used badly can make good audio sound awful.
Loudness maximiser is a powerful compressor so also consider with loudness maximiser and the compressor in my first option the destination of the audio. If it’s going to radio it’s probably going to run through heavy processing between the desk and the transmission so if you over compress at this stage it will sound awful once the station processing kicks in so use compresison as lightly as possibly at this stage.
If it’s internet radio or FM radion that doens’t run through heavy compression then you can afford to be a little more heavy with the compresison during your editing / mastering.
I would suggest the easiest thing to do for the future is to ask the talent to supply vocals to a certain level so the problem is eradicated as early in the process as possible and avoid possibly compromising the quality of the final audio through heavy processing.
Also, considering what processing the radio station uses. Chances are they’ll have a powerful multiband compressor running and a limiter too which will try it’s hardest to level out the audio once it leaves the desk and into the processing chain but it hits final destination, FM / DAB / Online. In that case it may be that you don’t need to anything at all as the station output processing could possibly do the work.
There’s lots to consider before doing anything so keep copies of your originals so if you don’t like how it sounds you have those to refer back to and try again with different settings on your various options.
Thanks.
These are recordings done a few years back at an FM station, so the voice is pretty solid.
I will try some of your suggestions.
IF the recordings are ok then you should be ok boosting the levels.
Hope you can get it sounding how you want it
