Studio Recording - Audiobooks - Sound Engineering

[G900Castle]

Hi there,

Curious, so when sound engineers and producers are working in the studio during live recording sessions with actors for audiobooks. When it comes to the editing stage, I understand that they need to edit out mistakes the actor might have made, but what other editing needs to be done?
Maybe normalization? Making sure all the levels are consistent?

Surely, if it's in a professional studio there will be no need to use compression/noise reduction.
I imagine plosives would be noticed straight away during the live recording, so the producer would tell the talent to repeat that line a little further back, head titled on an angle, etc.

Would love to know more!

Any input would be greatly appreciated!

[The Red Bladder]

You stick 'em in a booth and they and you have a copy of the script and they read out loud. I've become a fan of the MKH416 for this stuff as it can be two or three feet away. If in doubt and to save time, I use two or three mics and pick a winner afterwards.

Every time they make a fluff, you hit the marker button so that you can find it later. Then you edit out the fluffs, normalise it to -1dB or so and that's it. Put each chapter into a folder, bundle it all together and send it off (FTP) to whoever wants that stuff.

Sometimes a producer of sorts might attend, but that usually holds things up. Name-talent will know how to read out loud!

Sometimes a producer wants to listen via Zoom or Skype but again, that holds things up and if you are dealing with an experienced actor, they really do not need to be told how to voice the words.

Ad VO and ADR work will be attended by producers and directors and other similar animals.

[G900Castle]

Oh great! So it's as simple as that!

No need for corrective EQing, or de-essers.

Just normalization - Brilliant. Thanks for this!

[Hugh Robjohns]

Oh great! So it's as simple as that!

It can be...

No need for corrective EQing, or de-essers.

Now that depends mostly on the acoustic quality of the room, but also the choice and placement of the mic(s), and the nature of the talent's voice.

[G900Castle]

Oh great! So it's as simple as that!

It can be...

No need for corrective EQing, or de-essers.

Now that depends mostly on the acoustic quality of the room, but also the choice and placement of the mic(s), and the nature of the talent's voice.

Hi, I'd love to hear more.

Would some corrective EQing be used say if a voice talent sounded too 'boomy' - it would be implemented to bring it back a little?
And vise-versa for someone on the high-end frequencies, brighten up their voice a little?

[Hugh Robjohns]

Hi, I'd love to hear more.

Sure... but there's a practical limit to what can be taught over a forum.

Would some corrective EQing be used say if a voice talent sounded too 'boomy' - it would be implemented to bring it back a little? And vise-versa for someone on the high-end frequencies, brighten up their voice a little?

Yes, if necessary, and you can trust your experience and monitoring to provide a reliable reference from which to judge 'boomy' or dull.

[G900Castle]

Oh right, brilliant thanks for this!

All starting to make sense.
Truth be told, my friend has bagged himself an interview with his local studio, where they produce a lot of audiobooks (I'm super jealous) so we were both a little clueless about what editing techniques would need to be done if all the raw audio is going to be recorded in a professional studio. As he's been asked to explain in his interview what editing he would do for audiobooks. Think we were probably both overthinking that question, haha.

[G900Castle]

One thing, which I forgot to mention.

To make the audio consistent, is it best to Normalize it to your desired dB?
Or best to implement dynamic range compression?

Or even, I've heard some dynamic range compression first then normalize it at the end?

[James Perrett]

One thing, which I forgot to mention.

To make the audio consistent, is it best to Normalize it to your desired dB?
Or best to implement dynamic range compression?

Or even, I've heard some dynamic range compression first then normalize it at the end?

This totally depends on who you are recording - a good actor will control everything with their delivery. All the engineer needs to do is put them in a good room with a good mic and make sure they're recording everything. Other actors may require compression or post production fader riding to give the results needed.

The client should tell you the delivery specs they need but, when I last checked, BBC podcasts seemed to be sitting around an average level of -24LUFS.

[Hugh Robjohns]

... he's been asked to explain in his interview what editing he would do for audiobooks.

I imagine that's one of those interview questions intended to draw out the applicant's experiences and thinking, rather than expecting a definitive answer, since it would all depend on the voice talent's competence.

I've recorded sessions without any editing required at all, but others have needed lots of editing to remove fluffs, misreads, lip-smacks, table thumps and so on... Sometimes you can drop in individual words, but more usually you would replace a sentence or even a paragraph, particularly where the flow of the speech is important. Sometimes you might need to edit to reduce or expand gaps between words -- but beware accidentally removing breaths or creating a double breath!

[Hugh Robjohns]

To make the audio consistent, is it best to Normalize it to your desired dB? Or best to implement dynamic range compression?

Normalising doesn't make the speech level 'consistent', it just makes all the peaks the same... which is a very different thing! Normalising is usually about 'compliance' with a technical delivery specification.

Dynamic range compression (and/or limiting) is about creating a more consistent dynamic range, but it needs to be done with care and thought. A good voice talent might not require any compression at all, whereas Brian Blessed most definitely would! The nature and style of the reading will also determine whether some compression is helpful, of course.

Poor compressor/limiter settings can make the compression process obvious and distracting, though, and can also raise the audible noise floor undesirably and/or expose unwanted room colourations or background noises. It may be that some gentle fader automation in post-production gives better results, or maybe just some editing with level shifts to reduce occasional problematic syllables. For most broadcast-style podcasts and books I would probably be looking for 3-4dB of gentle 2:1 compression.

Or even, I've heard some dynamic range compression first then normalize it at the end?

That would be the sensible order of processes.

[G900Castle]

This is all fantastic!!

Thank you all for your input in this - really helped a lot!!