How To Review Your Final Audio The Audible Studios Way

Today we’re offering advice for producers and rights holders on reviewing their final audio for ACX. The steps we’ll outline can be used by a narrator before submitting their final audio to a rights holder, and authors can apply the same method before clicking “approve” to send the book to ACX for processing. We’ve got some great insight from the Audible Studios team, so let’s get right to it.

A Two Step Process

Audible’s editors listen to the entire book end to end, twice through, while following along with the manuscript. The first pass is called the edit pass, and the editor is mainly listening for and fixing technical deficiencies: sounds under words or “in the clear ” (between sentences), loud or unnatural breaths, mouth noises, plosives, pacing issues, and consistency of sound over the course of a long day of reading or between multiple sessions on different days. Audible editor Ashlee Harrison offers her advice on what to listen for when editing:

The most important thing to remember about editing audiobooks is to make the pacing sound natural (in regards to unnecessary or non-existent spacing). Something that I’ve learned really bugs listeners is unnecessary mouth clicks, and distracting noises that could easily be removed. Also, be sure to look out for cut or unnatural breaths. In some cases these things can be completely removed or simply cropped with a fade in to make it sound better.

The second, or “QC” pass focuses on the read, with an editor listening to make sure the narrator is voicing the words exactly as written. They’re also listening for mispronunciations, as well as ensuring that character names, place names, and voices or accents are consistent throughout the book.

How Audible Gets It Done

When it’s time to edit/QC, the issues listed above are either edited out or marked to be rerecorded (also called a “pickup”). When editing, it’s important to do so cleanly; that is, to surgically remove offending noises and keep the pacing consistent and appropriate by inserting clean room tone when appropriate. When marking pickups in the script, highlight the sentences before and after the portion containing the error. These surrounding sentences should be rerecorded as well. This will help ensure the editor is able to seamlessly insert the newly recorded audio into the original file. You’ll also note the particulars of the mistake, and where/when it occurred on a “QC Sheet.” This document is essential for ensuring that your list of corrections are organized and easy to understand. You can find a useable example of Audible’s QC sheet here.

Audible’s editors aim for a ratio of 3:1 on the edit pass and 1.2:1 on the QC pass. This means that a 10 hour book should take roughly 30 hours to edit and 12 hours to QC, though it may take longer depending on the subject matter, the language used, and the amount of errors made by the narrator. This strikes the right balance between attention to detail and the need to produce the title in a reasonable amount of time. ACX rights holders that are reviewing final audio can can focus on the QC pass, leaving the slower, edit focused listen to their producer.

With both the producer and rights holder listening in full, you’re bound to catch nearly all of the items in need of correction in your production. Your listeners will appreciate the attention to detail that produces a great sounding audiobook.

What are your secrets to a successful edit/QC? Tell us in the comments!

11 responses to “How To Review Your Final Audio The Audible Studios Way

  1. A note on breaths. People often ask how many breaths to edit out or leave in. And the answer is: it depends. More breaths should be left in for a dramatic (fiction) read than for an instructional read (nonfiction). Even in fiction, you should probably edit out big, gulping breaths, breaths at the beginning of a new paragraph or scene, or breaths that that disrupt the natural flow of the read. Lastly, be careful not to remove too many breaths. The read shouldn’t sound robotic, and humans do need to breath after all.

  2. Are the items listed above, also things needed to pass ACX approval for commercial sale? Thank you much! Am enjoying the learning process.

  3. Great article. Thanks. A nice trick I’ve learned (almost by accident) for editing in Reaper (for anyone who uses it) is to follow the folowing setup in edit mode:
    1. Take a small sample of the room ambient noise (in Reaper you can get away with microseconds worth as long as it doesn’t have a click or anything that will be noticed on looping).
    2. Copy that segment and paste it in a separate track.
    3. GLUE IT as is (it’s in the right click menu of the track) then drag that sucker (loop) all the way to the end of your audio length AND THEN A LITTLE MORE. It’s important to Glue it first or you will just get a looping of the track you copied, which will be more than just the segment you wanted. By glueing it all you will loop is the ambient segment you wanted. The reason to go “a little more” is seen in step 5.
    4. Change the color of this ambient noise track. I like yellow but go with whatever. It helps to differentiate it from the vocal track.
    5. Drag this track (don’t copy, just drag it) up to your vocal track. Because it is longer than your vocal track (even if it’s spliced or done in multiple-take punch-and-roll fashion), it will tuck behind your track. If you see faders, you didn’t make the track long enough (look both to the beginning and the end of the length). If you did it right, it’ll be just ‘behind’ your track.
    6. When you get to spaces between words on your editing now, all you have to do is splice and drag. Instant ambient noise behind your vocal. No need for fancy cut/paste functions.

    ONE WORD OF CAUTION: there is a small glitch with this trick. When you splice, you must make sure you have your vocal track highlighted, not the track behind it. You won’t really notice the track behind it being selected if that happens and you may end up just splicing out your ambient background rather than your vocal. It will become obvious though when your background ambient suddenly comes to the front. In that case undo and make sure the correct track is selected (the easiest way is to click right at the top edge of the track when the cursor turns into the volume adjustor for tracks…clicking then will highlight the track).

    NOW, having given this tip, I’m gonna say it makes between words easy to edit but darn if I’m not struggling with palate clicks that occur with my words (embedded). Would love advice for those.

  4. Very helpful article. My only advice is to use BOTH headphones (edit pass) and studio speakers (proofing). You’ll hear all of the tiny mouth clicks and other ambient noises with the phones; you’ll hear it the way the audience will hear it with the speakers. Also, when speaking in an accent, don’t worry TOO much if every single word isn’t perfect (I’m currently recording a book with ALL British characters, including the narrator) if it will disrupt the flow of the sentence. If it is intended mainly for American audiences, they can forgive small lapses (SMALL lapses!) to stay in character.

  5. Pickups:
    I go to the original file and record the corrected sentence over the old one, making sure that the new sentence matches the preceding and following sentence. I note the ProTools region number of the new sentence.
    After replacing all the incorrect sentences, I open up another program, GoldWave, and start a new file. One by one, I paste the corrected ProTools regions into this file, as a string-out, 5 seconds between each.

  6. That’s good information, thanks.

    I’ve run into a bit of confusion lately: Originally, I recorded completely ‘flat’ (No eq, no compression). For my last book, the author wanted to insert musical notes to differentiate scene changes within chapters. To make sure these notes didn’t become too tinny and potentially irritating for the listener, it occurred to me to roll off frequencies below 80 hz in my voice tracks and leaving the music alone before uploading.

    When I called ACX to discuss this, I was told to deliver the audio mastered and ready for retail. I thought that strange, since the final audio in my previous books was clearly processed and eq’d. Nonetheless I assumed that perhaps ACX had mastered my audio on my behalf, assuming I was unable to do so- and went ahead and applied eq and compression, and uploaded a ‘ready for retail’ finalized audiobook.

    When I heard the final audio, I was surprised to hear that my audiobook had been RE-processed, with a much smaller dynamic range and much thinner eq’d sound.

    So, my question: Do you want the uploaded audio to be delivered fully mastered, with eq and dynamics processing – or completely flat with no level adjustment whatsoever?

  7. This is a good information!
    You have given a number of effective tips on how to use the QC sheet. Thanks a lot!

  8. Pingback: Quality, Assured | Audiobook Creation Exchange Blog (ACX)

  9. Pingback: The ACX Author’s Audiobook Checklist | Audiobook Creation Exchange Blog (ACX)

  10. Thanks, I just found this post. I’ll have to try the QC sheet next time!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s