Processing Audio Files for Radio Broadcast & Archive??

[JKSAW]

Could someone explain to me the proper settings for processing audio for broadcast on FM radio, Internet, & Archive (podcast)? I have bee using Adobe Audition for a while, but this software has so many bells & whistles that I am overwhelmed. I know how to remove dead air, splice files, & use normalization, but I'm a bit cloudy as far as what processing settings I should use before saving to .WAV or .MP3. I'm most concerned with keeping the level even, as I have to occasionally deal with 78RPM records that were recorded before modern audio processing was invented. Perhaps someone can help me with this, as I didn't have the luxury of going to radio school.

[ryclark]

You may not haven't received any answers yet because your question is so broad. The typical answer would be 'How long is a piece of string'. It is very difficult to be specific without writing a whole book about audio. So just as a start here are a few suggestions. normally if you are supplying to an end user they will have specific requirements and rules as to what form and how their audio is delivered to them.

However for Archive purposes you need to do as little processing as possible. Don't worry about all the bells and whistles in Audition. You only need to use them when necessary for cleaning up or shaping audio for a particular sound. audition's help files should give you some basic information about all the included effects. But you may also need to do some studying of online tutorials and also search the internet for more in depth explanations of some of them.

For Archiving basically only normalise audio levels to make as much use of all the digital bits as possible, usually to about peak levels of say -3db. This preserve all of the available audio content and will allow any future access to the file to have a much information to work with so that a later user to can make their own mind up as how to process the audio. You should always store the audio as .wav files with at least 24bit 44.1kHz or 48kHz sample rates for the cleanest representation of the original audio signal.

Remember than that .mp3 should only be used for final distribution for listening purposes, not for storage, as it is a very lossy format. Requirements for other uses depend entirely on how the end product is going to be distributed and listened to by the end user. it also depends on the content of the audio material, whether it is music or speech and how much dynamic range is required at the far end of the chain.

[Bob Howes]

I have a slight problem with this because I'd have rather different answers depending whether the end result is for radio broadcast or simply archiving.

In the case of archiving I go along with ryclark.  Process as little as possible and nothing (like EQ or whatever) that actually alters the sound.  Yes to normalising (with headroom left as per ryclark's advice).  If the material starts as a wave, I might say to save it in the original sample rate/bit depth but there are also some arguments for going to 24 bit.  However, in the long term there are more devices that can handle 16 bit/44.1k sampling/stereo than can handle 24 bit.  As ryclark says, avoid mp3 as if it were the black death for audio.

For radio broadcast, it gets far more difficult for a couple of reasons.

First off, most broadcasters will have guidelines for how they want files delivered.  These can vary greatly--some want everything in MP3 and others have specific broadcast wave specs.  You really have to know where it's going.

Second, you mention 78rpm records.  At a very basic level, some broadcasters might want it to sound "old" like a 78.  Others may want it cleaned up as much as possible so it sounds like music rather than an antique.  In this case, the whole gamut of Audition processing (noise reduction, click and pop eliminator, etc etc) might be called for.

Sorry to be so vague!

[SteveG_AudioMasters_]

You might want to have a look at this site:

Guidelines on the Production and Preservation of Digital Audio Objects (web edition) | International...

[JKSAW]

Thanks for your reply. I know that my question is rather broad, but I wanted to get some idea of how to fix these issues.

I have been using .WAV files for broadcast, however the site I use for Archiving my shows (Mixcloud) only accepts .MP3. I've been using 16 bit. Should I be using 24 or 48 bit instead?

My biggest issue is that the volume level tends to fluctuate between songs. I'll go from one song with a lot of compression to a song that wasn't mastered to a 78RPM with no compression. I do use normalization, but it doesn't always work to keep the levels even thru the entire show. Not sure how to handle that.