Secrets of Voice Acting: Properly Formatting Audio Files

In an industry requiring even basic IT knowledge, you must be able to know how to format audio files in the proper manner.

This step is essential if you’re still a budding voice actor and want to become an independent talent with or without your home studio. So you can’t ignore the technical sides of audio recording and delivering  because sooner or later your clients will be asking questions about that or if you set up your own recording structure it’s obvious that you need to be familiar with this aspect.  But now worries it’s very easy to overcome this hurdle.

Which audio format is the best?

When formatting audio files you can choose between .wav (Windows) and .aiff (Mac), no substantial difference between them really. What matters most is the sampling rate and the resolution or bit rate

Sampling rate means the number of times that the audio card takes a sample or register the sound from a continuos signal, therefore the higher the rate the more sound it can capture. Most used sampling rates are 44.1 kHz, 48 kHz, 88.2 kHz, 96 kHz and 192 kHz. For IVR phone messages the sampling is even lower 8 KHz, simply because the audio files will be very light and easy to carry on telephone lines. For voiceover recording we recommend 48 KHz which is the industry standard for any HD project. 44.1 KHz is still widely used because it’s still the sample rate of CDs that were born back in the early 80s, but in the age of DVDs, Blueray, 4G Mobile networks, you would be more professional recording and delivering in 48 KHz, no question about that. But hey clients are kings, so if they want a lower sampling so be it, but then record in 48 Khz and convert into the desired format.

Resolution

Also called bit depth or bit rate, you may choose from 8 bit to 16 or 24 bits. Bit rate relates to the amount of information that the audio card captures from a particular sample, the higher the rate the better the sound but heavier the resulting file will be. For voiceover 16 bit depth with 48 Khz sampling rate is the perfect combination to capture your voice’s dynamic range. Lower than 16 could mean unwanted digital noise or grain that is audible in the pauses. So avoid 8 bits and go for 16 bits or higher. Again if the size of files is important, like for mobile application 8 bits can be used, but then as an exported format, not as an recording format.

I don’t know why but most of us voice talents are not good at maths, and we are using maths when we record. Not a bad idea to have a look on what is digital sound, how our voices turn into thousands or millions of binary code lines.

If you don’t like maths don’t be scared.

In his voiceover classes Aliso Creek has explained it very well with layman’s terms at this post titled Bits and Bytes, Understanding Digital Jargon for your Home Voice Over Recording Studio. 

Once you record, I strongly recommend you deliver a normalized file.

This means that the master audio in .wav or .aiff should be treated but with a gentle processing method that consists in these steps:

1) Once you have recorded and done a rough edit (just removing the errors), you copy that master file and you paste it in a different track, the post-production track

2) In this post-production track, we use the plugin present in most audio softwares called Vocal Leveler (you have it installed in any ProTools for example). This plugin just reduces the dynamic range without compromising the audio quality, it just avoids the peaks that could result in distortion. It’s a very gentle compressor. FYI a real compressor should be avoided.

3) Once you have your file vocally leveled, you apply a the Normalizer plugin (also included in the Audio Suite package in ProTools and present in most audio softwars). You will set the normalizer at -2db and then you render.  Provided that you use the right sampling and bit rate (we advocate 48,1 KHz), he result will be perfect for any application: Internet, HD usage and even broadcast.

4) Only then when you have normalized you can make a final edition, that removes unwanted breaths, clicks and crackles.

 

What about mp3 ?

By definition this patented encoding system created by Fraunhofer Institute, Germany is a very compressed format. Its smart algorithm just removes many unnecessary components of the sound without compromising the audio quality. This format is ideal when delivering your demos. It can be very light so you can deliver the audition samples through simple e-mail.  Still we recommend you record in .wav or .aiff and then convert into mp3 with the highest resolution possible, this will allow to reduce the size of your file at least 6 times.  Gilles Arbour argues that the gist is that mp3 is the best audio file format for demos since it is compatible with most players.

 Alexander Abbott has described the process of making demos in a nutshell:

• Record yourself then open an audio converter software. Most popular ones that come with a price are Sound Forge, Switch Plus and Audials Tunebite. Free softwares for audio files formatting include Audacity, Pazera, Zamzar and Super all available for download online.

• The thing with converting files is that you only need to convert when the end product of what you have recorded isn’t compatible or you want it converted into the most popularly used MPEG files with a mp3 extension. This can be played by any music player and most users of Mac or PC’s either have Wav or Aiff extension file names so they still convert it to mp3 for it to play better in most media players.

• Open or import files you want to convert and make sure your recording is just 1 and a half minute long to not bore the judges.

• Conversion time usually lasts according to how long the file is or how many files are you converting, so lesser number of files equal lesser time.

• Rename your files in the usual way after the files were exported, right click to rename.

• Save it on your desktop then get a blank CD or DVD to burn the track into, then label it with your information. But now the trend is to just store in a secure server, no need of physical support.

I’m sure there are voice talents who like maths and are passionate about the technical side of this profession. If you are a geek don’t hide, please share below the useful things you have learned when recording and delivering audio.