Audio editing advice for Lecture
-
- KVRer
- Topic Starter
- 1 posts since 17 May, 2015
Hi,
I'm working on a project where I will be touching up audio from dozens of lectures recorded with the same speaker and equipment.
The goal is to lower background humming, raise the speakers voice, reduce deep bass in his voice from his amplification set up and lower extreme highs in the audio...laughing/coughing near the mic or the speakers high points.
I have sony noise reduction, wavehammer and a host of other plugins, and can get the audio to sound really nice but it seems like I'm working in circles at times, repeating some steps over and over because one effect seems to counter act some of the effect the last plugin used.
I'm wondering if anyone has a typical protocol or plugin combo that would more efficiently achieve my stated goals.
Thanks for any info
I'm working on a project where I will be touching up audio from dozens of lectures recorded with the same speaker and equipment.
The goal is to lower background humming, raise the speakers voice, reduce deep bass in his voice from his amplification set up and lower extreme highs in the audio...laughing/coughing near the mic or the speakers high points.
I have sony noise reduction, wavehammer and a host of other plugins, and can get the audio to sound really nice but it seems like I'm working in circles at times, repeating some steps over and over because one effect seems to counter act some of the effect the last plugin used.
I'm wondering if anyone has a typical protocol or plugin combo that would more efficiently achieve my stated goals.
Thanks for any info
-
Winstontaneous Winstontaneous https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=98336
- KVRAF
- 2344 posts since 15 Feb, 2006 from Berkeley, CA
I've done a lot of live performance video/audio editing...the fact that it's the same speaker/equipment makes your job a lot easier.
For this kind of work I find spectral tools (I use Adobe Audition) essential...being able to see noise patterns and address them individually is so helpful. I evolved a basic workflow:
- listen while viewing spectra in realtime
- drop markers to note what to cut and where (be sure to label markers!)
- crop clips to usable sections
- lay out clips on timeline
- apply high/low pass filtering to address rumble, AC noise
- address obvious noises with de-essers, gates, spectral & adaptive noise reduction
- get individual clips close to same volume with clip gain adjustment/automation or normalization
- apply EQ/reverb/etc.
- compress/limit
- if necessary - multiband compression/dynamic EQ
- fades
- switch between headphones and several sets of speakers (if possible) throughout the process
Basically, cut and trim as much as possible before you boost/EQ/compress, in order to minimize noise and artifacts. Don't expect processing applied to a whole channel/track to work unless all the individual clips sound as good as possible on their own and are of equal loudness. Hope this helps.
For this kind of work I find spectral tools (I use Adobe Audition) essential...being able to see noise patterns and address them individually is so helpful. I evolved a basic workflow:
- listen while viewing spectra in realtime
- drop markers to note what to cut and where (be sure to label markers!)
- crop clips to usable sections
- lay out clips on timeline
- apply high/low pass filtering to address rumble, AC noise
- address obvious noises with de-essers, gates, spectral & adaptive noise reduction
- get individual clips close to same volume with clip gain adjustment/automation or normalization
- apply EQ/reverb/etc.
- compress/limit
- if necessary - multiband compression/dynamic EQ
- fades
- switch between headphones and several sets of speakers (if possible) throughout the process
Basically, cut and trim as much as possible before you boost/EQ/compress, in order to minimize noise and artifacts. Don't expect processing applied to a whole channel/track to work unless all the individual clips sound as good as possible on their own and are of equal loudness. Hope this helps.
-
- KVRAF
- 7540 posts since 7 Aug, 2003 from San Francisco Bay Area
It's actually way easier then that to remove background noise in Audition. Just find a spot where no one is speaking, select it, and capture it as a noise print. Then you can use the noise reduction tool to remove the background noise based on that sample.
Incomplete list of my gear: 1/8" audio input jack.
- KVRAF
- 4278 posts since 6 Nov, 2009
Very good advice. Sometimes noise removal and EQing take too much out. Blending the two together would sound nice.Scottex wrote:Parallel processing will help you. Process one copy and blend it with the original version.
you can duplicate track or use busses and routings, as you like. Hi