How to Edit a Dramatic Work

So, you think you might like to take up editing audio files together to create a dramatic work? Here's some information about how to go about carrying out this fun and rewarding LibriVox activity.

Editors take various approaches to editing dramatic works and of course everyone is free to use whatever method works best for them. This page sets out my own approach (using Audacity) and may work as a starting point for learning about audio editing.

You can wait until all roles in a project have been completed before starting to edit, or you can edit as each act/chapter's roles are completed. The latter approach spreads the work of editing out so you're not having to spend a lot of time on it all at once (my preferred approach, as I like to have variety in what I do on LibriVox).

Below are the steps to follow:

Download Files for Editing

Download all the completed/PL OK files to your computer and save them into a separate folder for each chapter/act. The files you download will be in .mp3 format, but you can just open them in Audacity and save them as .aup3 files. No conversion necessary. (You may want to go back and delete the original .mp3 files after you’ve finished editing, just to reduce clutter.)

Use the Narrator/Stage Directions File As Foundation

Open the narrator/stage directions file in Audacity. Immediately save the file as an act or chapter (for example, "Act 1" or "Chapter 12") using your project’s file naming convention (e.g., wouldbegoods_12_nesbit_128kb or hamlet_1_shakespeare_128kb). Listen to some of the file using headphones to determine whether it needs noise reduction or overall volume adjustment. Apply noise reduction and/or adjust the volume as needed. Make sure not to over noise reduce a file; you want to remove background noise without adding the metallic-sounding "artifact" that is produced when a file is noise reduced too much.

While you can apply any necessary processing to each of the characters' files at this stage as well, I prefer to do the individual files as I go along, since they’re usually short. That includes making sure there aren't any significant changes in volume between characters.

Start Editing: Two Methods

Here are two ways to tackle editing.

Open Files As Needed and Paste In Lines

Listen to the narrator/stage direction file until you come to the first place where one of the characters speaks. Open that character’s file and find their first speech. Select and copy the speech, and minimize the file.

Paste that speech in about half a second after the end of the narrator’s last words. If the speech is followed by the narrator speaking again, close up the space that follows the pasted-in speech to allow half a second (or however long you judge to be a suitable silence at this point), and go on to the next one. If there’s a significant change in background noise or volume between the narrator and the role, apply noise reduction or amplification/de-amplification to each file or speech as needed.

When I first started editing, I'd paste in only one character's line at a time then listen to each narrator/character/narrator section separately. Now that I'm more comfortable, I edit together three or four different characters' lines before going back and listening to what I've just done. I try not to edit more than a page/page and a half of material before going back and listening to it, just in case I’ve pasted something in the wrong place and have to fix it (much easier to do if it's just two or three lines).

Rinse and repeat until the whole act/chapter is finished. You’ll have a lot of files open by the time you finish the act/chapter, so find whatever method works best for you to keep track of those so you can easily find them. I'm on a Mac so minimizing the files neatly lines them up at the bottom of my screen in the order in which I opened them (or last used them).

When you've finished editing in all the lines, listen to the entire file all the way through to make sure it sounds natural (listen for any thumps/clicks that might have been missed earlier) and adjust the silences between lines if necessary.

Open Everything in Two Files and Paste from One to the Other

Open the foundation (narrator/stage directions) file, then import all of the other characters' files into another separate Audacity file (you can just drag and drop the files, and they'll each appear in a new track below the preceding one). Save that file as "Working files, Act/Chapter 1" or something similar. This approach is probably easiest to do if you have two monitors, one for each file, or a wide monitor.

Each track in your "Working Files" will have the full file name at the top. If you like, you can rename each track with just the individual character's name (click the three dots next to the file name at the top left side of the track. A drop-down menu will pop up, with "Rename Track" at the top. Click that, and type the character's name into the dialogue box that comes up.)

Tip

Move the tracks up and down within your "Working files" to place characters next to each other who are speaking to each other in a particular chapter/scene; this makes editing scenes with two or three characters much easier, rather than having to scroll up and down to find each character's tracks.

Listen to each character file individually using headphones (click "solo" at the head of the track) and determine if noise reduction is needed. Apply noise reduction and/or adjust the volume as needed to each character track individually. Aim for making the volume approximately equal for all tracks before beginning to edit. (You can use the "Loudness Normalization" effect in Audacity or do each one manually, using Amplification.)

Following along with the text, cut and paste each character's lines individually into the main file as outlined above. Make sure to pay attention to timing (getting the right amount of silence between lines).

When you've finished editing in all the lines, listen to the entire file all the way through to make sure it sounds natural and that there aren't any thumps/clicks missed earlier, and adjust the silences between lines if necessary.

Other Editing Tasks/Questions

Here are some other tasks/questions you might come across when editing.

How Do I Fix Clicks in the Files?

That’s an easy one. The De-Clicker (link at the bottom of the first post), an Audacity plug-in, is quite effective in removing clicks in files without affecting the sound of the recording.

To use it, highlight a section containing the offending click — it works best to select more than just the sound you want gone so the plugin has some kind of baseline of what is "normal," and can recognize what to remove. Apply it only to those areas that seem to have clicks in them, as it may overly soften the sound of your p, t, k, and b sounds.

What About Background Thumps?

Those are harder to tackle. You can try to remove them if you can isolate the sound (or if it's in one of the silences), but if it's behind speech, you'll likely have to go back to the reader and ask them to re-record the phrase it happens behind.

What If I Find Missing Lines While Editing?

If you come across a missing line while editing, mark the space by leaving/adding 5 seconds or more of silence (copy some room silence from later in the file rather than generating absolute silence). You can also add notes to any place in the track by selecting some of the silence and clicking "Edit > Labels > Add Label at Selection" and typing in a description of what needs doing so you can find it easily later.

What If the Reader's Not Available Anymore and Something's Missing?

This is a bit trickier, but not impossible to fix. Consider first of all whether the missing word (if it's just one or two) is essential or whether you could just leave it out. If it's absolutely necessary, do a search through the text to see whether that particular reader has read the word(s) elsewhere in their role. Listen to the other instance(s) and decide whether you might be able to copy and paste it in where it's missing. This will likely only work with a word or two.

You might be able to use this approach as well if a reader has read a word correctly in one place, but incorrectly elsewhere.

If there's no replacement word available in the reader's voice, you might also consider asking your narrator to read the entire sentence in which the missing word appears. The narrator likely has a lot of lines, and perhaps "speaks for" characters indirectly elsewhere, so they would have the least surprising backup voice. You might use this option for entire missing lines as well.

How Do I Combine Multiple Voices Together?

Sometimes you'll come across a word/line that is spoken by several characters at once (e.g., in crowd scenes). Here's how to merge those.

Copy the first character's line into the main file, then select "Track > Add mono track" to add as many new tracks as you have voices to combine (minus 1), copy and paste each character's line into a new track, adjust them so they align properly below each other, then listen to them all combined to make sure you like the sound before rendering them together. You might want to adjust the silences slightly before/within/after each line to allow for slower/faster speakers and to make sure it all sounds natural.

Once you're happy with the sound, select All (Cmd/Ctrl-A) and click "Track > Mix and render." That combines all of those extra tracks into the main one and merges the voices together (once you’ve done this, you can no longer adjust them individually). Then, if necessary, select the segment containing those combined voices and reduce the volume a bit, as merged files tend to get louder when combined, even if they were individually just fine.

Can I Cut Stage Directions/Narration?

Yes and no. Although LibriVox policy emphasizes including "the whole text and nothing but the text" in finished projects, stage directions in plays may be left out, if their sense is conveyed by the speaker (for example, leaving out "John laughs" if the character has recorded some laughter at that point). But for Dramatic Readings, all text must be included as written by the author.

Can I Add Music/Sound Effects?

No. In keeping with LibriVox policy on sound effects and music, please do not use any in dramatic works. If a character sings a song in a play, it's fine to have the reader make up a tune (as long as it's public domain) and sing it without accompaniment for the recording.

Alternatively, the BC could come up with a melody for the song and have the readers perform it. Copyright issues can be tricky to handle with sound effects and music, so we try to avoid taking on these issues whenever possible.

See the official policy here.

---

See also "8. Suggestions for editing a dramatic work" in the "Good practice guidelines for dramatic works" post on the Forum.

If you prefer visual to written instructions, see the YouTube videos How I Edit Plays and Dramatic Readings by Phil Chenevert, and LibriVox Zoom Meeting Oct 11, 2020 by Todd (starting at about 23:07), who both set out clearly how they edit dramatic works. (Links to these and more videos can also be found on the LibriVox Wiki, under Instructional Videos.)

See Also

• An Introduction to Dramatic Readings and Plays for New Readers
• How to Create a Dramatic Reading Script
• An Introduction to Proof-listening a Dramatic Work
• How to Become a Dramatic Reading/Play Coordinator