Landed my first spoken word recording gig... any advice?

Are you just the talent or are you doing production on it also? Workflow can change a lot depending on your role.

Ahh, gotcha…I misunderstood. I’ve done some audiobook stuff and what I’d recommend is:

- for your signal chain I’d focus on eliminating sibilance and plosives, and having a nice full range signal. I don’t personally worry too much about how colored it is, but I am worried about repeatability. Can you easily reproduce the sound? That might mean leaving any cool, esoteric pieces out of the chain. Also, like most things the mid range is super important. Podcasts, audiobooks etc are often listened to on earbuds, in cars, on laptop speakers etc. To translate well you need a nicely detailed midrange.

- leave any noise reduction off, do that in the mix later.

- record it as close to finished as you can. That way sorting through hours of dialogue isn’t needed.

- either edit as you go, having the artist punch in in real time (as needed) or keep dropping markers with extensive notes on which parts to edit out. Honestly, I think punching in as go will save you hours (possible days) of time.

- arrange your sessions by chapter, stanza etc whatever makes sense for the project. Don’t track the whole thing in one session. If you have to go back later and punch in a line on the second paragraph of chapter 3 etc, it’s way easier to navigate this way.

- check to see what the distributor wants as far as file format and delivery. Maybe they want the audio to peak at -10db, maybe they want a specific rms target. When mixing, just make it sound good and then normalize the entire project to their requested format.

- don’t underestimate how hard it can be for the talent to “perform” the poetry. Everyone thinks voice overs and spoken word is easy, and then they stumble over every word and it becomes a total shit show. Voice acting is a real talent/skill. You might find yourself having to couch the talent much the same as you’d coach a singer.


Anyway, that’s all I can think of off the top of my head. Hope that helps.

Would you recommend sending him back with "dailies" (not sure what they're called in VO) or does that just risk having him overthink the whole thing? The idea here would be that we can start each session with any punches that need to be done on the previous session.

Bad idea? Good idea?

There's a case to be made that redo-ing everything is just more studio hours. I feel gross even typing that because it seems kind of unethical or something, but isn't there a case to be made that if someone wants to spend twice as long doing something, more power to them?

Feels kind of wrong to me though. I don't like the idea of milking a client.

yeah, that's it.  if they have the budget and insist on "fixing" everything then let them go.  Maybe they do know better.  As long as you are not the one suggesting it then you are doing what the client wants/needs.  And if it sounds good, let them know it's good.

Never would have thought of that. Good suggestion, two mics... easy enough.

“This is the BBC. Here is the news.”

I'm a bit surprised no one has mentioned these 2 very important things :

#1 - Preferably use NO EQ, and use a chain that is very "standard" across the country / world. It's very common for fixes, pickups or little adjustments in other places of the world - as in not your studio. The mic should be common. A U87 is very standard, and use a mic pre (API or NEVE) that virtually every studio has a variation of and you'll be safe. If you need to EQ, do it in the DAW and use an industry standard EQ and notate the settings.

#2 - ROOM TONE!!!!!!!!!!!!!!!! Wherever you have a recording take place - voice over booth, main room, out in a lounge, wherever - RECORD ROOM TONE!!!!!!! Yes, I was shouting. If you don't know what this is, you've probably never had to edit dialog. This is a critical part of things, and make sure you slate it - "vocal booth, U87, room tone" and record 30 seconds or so. And yeah, anywhere you record, record room tone. The editor will curse you if you don't.

Another note - make sure you monitor back on something where you can hear sub bass so you can know FOR SURE you have no "LF pops" that you're unaware of.

Usually it's the background noise of the room that can be layered in to lengthen edits or fill silences where needed.

It always sounds weird to me when podcasts have a hard gate in between phrases/pauses, because you can hear how unnatural the “total black” silence sounds compared to the ambient room tone.   

Roomtone is good to have. But you can also "create it" these days or almost remove it entirely. I've done both for dialog. And if your doing a constant record you'll usually always have long pockets of room tone all over.


Two mics isn't bad, especially if its dynamic material. Or you can multi the mic and have it going to two pre amps with different gain. I do that for production audio sometimes.