I do a lot of recordings for long-form dialogue like audio books and industrials. For a trained professional actors it will be about 3 hours or recording per finished hour and then 2 hours of editing per finished hour. So 1 hour of dialogue for an expierenced actor & editor would be 5 hours or work total (3 recording + 2 editing).

I've also worked with a lot of less skilled non-union actors and that can vary a great deal. That's going to depend on the quality of the performance you want and the skill of the person. At the extreme I once spent half and hour recording the word 'wow' for a talking bottle of car-care product. On the other hand they can sometimes be just as good as a professional actor. But to give an average I would say 5 hours recording per finished hour and once again 2 hours of editing per finished hour.

Typically we budget about 100 lines per hour - given 2 maybe 3 takes per line.

If the lines are story driven then the delivery can take some work to nail down. If the weight of the lines have to portray a certain emotion or will drive the animation, more time can be spend on single lines - like the 'wow' example. A branding word, or phrase will get a high priority and many takes to achieve the vision.

Typically if you have a lot of dialog, you want someone from the client to be present - to help provide direction rather than guessing what they want and then having to re-book studio and or actor again.

If you have a lot of dialogue, it may also be worth splitting it into multiple sessions so that you can take the feedback and apply it to retakes if necessary in following sessions.

We have done some character based games where most of the dialog is callouts 2-7 words only, we average over 1000 individual takes which works out to about 300-ish lines in 2hours of recording and so the editing takes up to about 8-9 hours after that to name and find takes by an editor.

By the way, I highly highly recommend this piece of kit

https://auphonic.com

It's a batch processor which takes a lot of guesswork levelling dialog.

I run it both before processing and mastering to get my VO to a consistent level across the session, and then once more after processing and mastering to ensure it adheres to the final dialog levels we wish to have in game.

Thank you, that was actually quite helpfull and to the point :-)! Thanks. We will do the editing ourself, so we can cut the time at this. I suppose there was someone from the client side so he could immidiatelly say if he likes it or not? Or was it a 3 hours recording (with an union actor) for one hour dialogue and only after that was it sent for feedback from client and additional retakes would have to be done, if yes, how much time did the additional retakes (after client feedback) take?

(Also for the laughs, how is it possible to record half an hour a simple word "wow" from someone claiming to be voice actor (profesional or amateur)... :-)?? )

It's pretty rare for the client to be there because they don't really have the expertise to tell if something is good or not in the raw state like that. They only know if they like it when it's done. Usually it's a creative director who will either be in the control room or on a phone patch. Depending on how particular they are and/or how much they want to throw their weight around it can be much faster or much slower. On the client side they don't usually correct things piece-meal when they get the finished product because they've already approved the pitch and they've already listened to the auditions so they know what they're going to get. If there are pickups you need, they'll be very easy in this medium because you're not hard locked to time. As for the 'wow' example it was more the creative trying to be assertive with a new account rather than a actual problem with the performance.

Make sure to get someone talented, listen to their demo reel and make sure they can really do this. Also make sure they're set up right, because if it's recorded bad, it's going to sound bad. Most voice actors will have a home studio built out or at least a whisper room. Also make sure they have a decent microphone, if USB is mentioned, move on (c: You don't want to waste a bunch of time trying to polish a turd.

Finally I would recommend against automatic stack services like Auphonic. Editing voice over is the easiest type of sound engineering to learn and even if you're brand new at this you're going to be able to do it better than an automatic process. They also normalize which is a big no-no in the world of voice over, spend some time learning the basics and you'll be very glad you did.

Thanks for feedback/tips tricks, keep them comming :-).

Groovy one:

You wrote:

"Typically we budget about 100 lines per hour - given 2 maybe 3 takes per line."

100 lines per hour with 2-3 retakes in the end translates to how long final voice acting output? 30 minutes?

This averages to about 12 seconds per line. with 3 takes one after another then you get about 20 minutes of final audio.

So if the actor is able to nail the performance, delivery quickly in 2 takes you get more.

If the average line length shorter then you get more... etc.