Useful Links

• Skywind Pronunciation Guide

• Skywind Accent Guide

• Filecutting Second-Pass Guide

Prerequisites

Software

You will need three pieces of software for filecutting:

1. Audacity, a free audio-editing program.

2. LibreOffice, a spreadsheet software. We recommend LibreOffice in particular (even if you have Excel) because of the toolkit linked below. This will make filecutting much easier!

3. 7-Zip, a program capable of handling .zip files.

Accounts

Additionally, you will need:

1. A ClickUp account. This is completely free. An invite to the Skywind VA board will be sent once you've completed training. Miscellaneous ClickUp information can be found here.

2. A Discord account. Make sure you have joined the Skywind server: https://discord.gg/skywind 

Toolkit for LibreOffice and Audacity

The wonderful Krisscracka, Slaynot, and bradyjam have developed a toolset to make filecutting easier.

The latest release can be found here: https://gitlab.com/Slaynot/filecutter-toolkit/-/releases/1.4

Information on installing and using the toolkits can be found here: https://gitlab.com/Slaynot/filecutter-toolkit/-/wikis/home

What is filecutting?

When voice actors submit their recorded lines, the lines are all in one big file. The job of the filecutter is to take this big audio file and create a "label track" that assigns a filename to each segment of the audio submission. You won't actually be creating the individual files yourself. That will be done automatically using your label track after our SFX department has mastered the track. In addition to assigning filenames, you will check for pronunciation and script accuracy and decide what line to use if a VA submits two takes of the same line.

So, your task will be to download a VA's submission and script, create this label track, and make note on the script of any lines that were missing or need a retake.

The only requirement is that you try to get your assignments done within a reasonable timeframe. Try to make consistent progress, even if it's slow. If you can't finish your assignment, tell us what progress you made, and upload your files so someone else can complete it. 10-20 lines per day will finish a small or medium card at a good pace, and 30-50 lines per day is a good pace for longer cards. Be careful not to burn yourself out. If you find yourself making slower progress than you hoped, especially on a longer card, ask for a second filecutter to divide the work with.

What do scripts look like?

Below is an example of a Skywind script. The first row contains information about the character(s) being voiced. Things like names, race, etc. The second row contains some useful links. Then we have the script itself, where the third row contains the headers, and the following rows are the VA's lines.

Column 1: The filename for this segment. This is what you will label the segments as.

Column 2: The speaking NPC.

Column 3: What the player asks about or does to get this NPC response. This may be blank for things related to combat, detection, or idle NPC behavior.

Column 4: The line the Voice Actor reads.

Column 5: The facial expression the NPC will have in game. Please ignore this column, it should not be taken as an acting instruction.

Column 6: Any acting notes for this segment. This includes pronunciation guidelines for certain common words. The rest can be found in the Pronunciation Help (also linked in the script file!)

Column 7: Where you will leave your notes!

Instructions

Once you have been assigned a card in ClickUp, you can begin filecutting. These instructions assume you have the filecutter's toolkit installed and configured according to Lacuna's tutorial video found at the top of this page or here.

Setting Up

1. Download the most recent script (XLSX file) and the audio files (usually two WAV files, one spoken, the other combat) from the ClickUp card. Typically combat lines come at the end of the script, so start with the spoken file.

2. Open the spoken WAV file in Audacity. If you have an Audacity shortcut on your desktop, you can do this by dragging the WAV file unto your shortcut.

3. Before we begin labeling the file, we need to check a few things. VAs are supposed to have 20 seconds of room tone at the beginning of each file. Room tone is the ambient noise of the recording space when the VA is not speaking or making any noise. There should also be room tone in-between lines, and not dead-silence. Room tone is usually visible in Audacity's Spectrogram view. We can switch to Spectrogram view by left-clicking the track title and selecting Spectrogram. 

Room tone will usually look like a small line of pinkish purple at the bottom of the audio track - if it is black, with no color at the bottom, the room tone may be missing. If it looks black in Spectrogram view and you're still not completely sure if there is room tone or not, you can select the silent section, go to "Analyze" --> "Contrast...", then click "Measure Selection" to the right of "Background". Any result besides the word "Zero" means that the room tone is there, and you're good to continue. If you do find that the space at the beginning (or especially between lines) is totally black, and the "Contrast" result is the word "Zero", ask the voice actor if they have a version of their lines saved that still has the room tone.

  4. We will create our first label for the room tone. Select the room tone at the beginning of the track, or any period of room tone that is at least 5 seconds long and ideally 20 seconds, and press CTRL+B to create a label. Type "_ambient", then press ENTER.

  5. We will not be working in Spectrogram view, so we can switch back to Waveform by left-clicking the track title and selecting Waveform. We can also switch from a linear (amplitude) view of the waveform to a logarithmic (dB) view. This makes the waveform a lot "bigger" and easier to see. This also makes quieter noises that wouldn't show up in linear view visible. Even in logarithmic view, you can't see everything, but it makes it a lot less likely that you'll miss really quiet sounds. To change to logarithmic, right-click the scale to the left of the track and select Logarithmic (dB).

  6. Open the script file you downloaded from ClickUp.

  7. Save the script, renaming it with "Filecutter" at the beginning. (i.e. If the script file is "Training_Script.xlsx", save it as "Filecutter_Training_Script.xlsx").

Filecutting

1. Make sure you are listening with headphones and not speakers. The audio will usually be much quieter than anything you'd normally be listening to, so you'll need to turn your volume higher than normal, sometimes a LOT higher. Some audio issues are quiet and hard to hear, so turn your volume up, but not so much that it's uncomfortable, just louder than normal. This makes it less likely you will miss those quieter issues.

2. The lines should be in the same order as the script. If they're not, you can search the script for the line you're hearing. Do not reorder the script to match the audio or vice versa.

3. Select the first section of audio and take a listen. While you listen, read the script and make sure that the voice actor has followed the script exactly, with no deviations.

4. Lines should have 0.3 - 0.5 seconds of silent room tone around each line. It's fine for the labels from two different lines to overlap and share the same stretch of silence. DO NOT edit the audio to add more silence! The labeling macro automatically adds 0.5 seconds around the line, but you can adjust it down to 0.3 seconds as necessary.

5. Label the line appropriately in LibreOffice. The labels, their macros, and descriptions are listed below. This will correctly color code the line, enter templates in the Filecutter Notes, and copy the filename of the line. If the line has a problem, do not label it in Audacity. Only give it a label if it doesn't need a retake. The video version of the filecutting guide has examples of the problems you might run into - this text guide is helpful as a reference, but the video guide should be your primary source because it has audible examples of these issues.

   6. If the line is perfect: in Audacity, select just the line with no silence before or after the line.

   7. Press ALT+A. This creates a label with 0.5 seconds added on to each end of the audio we had selected and pastes the filename we just copied onto the label.

   8. Repeat steps 3-7 for each perfect line until all are marked and labeled. (Again, do not label lines that require a retake.)

   9. Select the entire label track by clicking in the space under the Labels dropdown.

10. Click File in the top left.

11. Select Export Other.

12. Select Export Labels.

13. Name the label file "Labels_[The exact name of the WAV file]". If the WAV file you worked on is "Synnolian Tunifus (Retakes - Normal).wav", your label file should be "Labels_Synnolian Tunifus (Retakes - Normal).txt".

14. Save the file somewhere you can find it later.

Combat

1. Combat should be as convincing as possible. We need real shouts and screams to be convincing. If they VA is not really shouting or screaming, or sounds unconvincing, that line should be marked for a retake.

2. Open the combat WAV file.

3. Check the room tone with Spectrogram view and label it.

4. Switch to logarithmic view.

5. Listen to each line and label appropriately just as you did for the spoken lines.

6. Unlike spoken lines, combat lines can have as little as 0.1 seconds of room tone around each line, but there must always be some room tone.

7. Be sure to keep an eye out for any clipping audio. Clipping is caused when a VAs hardware is overwhelmed by an loud noise and distorts, usually because their gain level is set too high. Any clipping should be marked for a retake. You can turn on "show clipping" in Audacity by clicking View in the top left and clicking Show Clipping in Waveform. This is helpful, but it's not 100% reliable, so listen closely for distortion. If the waveform looks flat at the top and bottom, instead of having a smooth curve like normal, that's a giveaway that there is clipping even if "Show clipping" does not reveal it.

8. Export the label track just as you did for the spoken file.

9. Don't forget to save and close your script! It's a good idea to use the Alt+X macro to make sure you didn't accidentally skip any lines.

Finish up

1. Select the filecutter script and all of the label tracks.

2. Right-click.

3. Under 7-Zip, select Add to archive...

4. Name the archive following the naming convention found here.

5. Click OK.

6. Attach the created 7z file by dragging and dropping it to the attachment section of the appropriate ClickUp card.

7. Leave a note saying that you've completed your labeling. It's a good idea to remind the VA to wait for a script before recording their next round of retakes.

8. Remove yourself from the card and remove the "Filecutter Assigned" tag.

9. Don't forget to click "Unwatch" under the bell symbol above the comments to stop receiving notifications.

10. Move the card from the first-pass column to the second-pass column, or from the second-pass column to the pre-retakes update column, by clicking on the status icon and selecting the appropriate status.

11. You're finished!

Exporting Audio Clips

1. Select the portion of the audio that you want to share by left-clicking and dragging along the timeline.

2. Click File in the top left.

3. Click Export Audio...

4. Set the Export Range to the Current Selection.

5. Change the filename before you save so that it doesn't overwrite the original WAV file.

6. Then click Export.

7. You can then drag and drop the exported WAV file in the Filecutting Discord channel and explain what you're concerned about.

Tips/Reminders

• Do NOT delete or edit any audio, only label segments using Audacity's labeling tool.

• Please leave friendly and helpful acting notes for our voice talent. You are their acting coach! A good sensible judgement of acting value and skill is needed for this. Use your best judgement and feel free to ask other filecutters for a second opinion. Quality is very important to us here at Skywind!

• Trouble with shouting lines? Don't worry, you are not alone! Be sure to look to the label name itself and the mood of the character in the script for hints! For example, the line for "MorroDefau_MDQPowerAttack_0004A0F7_1" has no prompt, but the filename tells you that it is for when the NPC is power attacking.

• Don't be afraid to tell things as they are. You are the last bastion of quality after all! If a VA is not convincing enough, report it! This is especially important for combat lines.

• If you are unsure about a pronunciation and don't see it on the guide, feel free to ask on Discord! You can also search the Discord to see if it has been brought up previously.

• If you have any questions at all, feel free to ask on Discord!

What Next?

After you've cut a few cards, you should be ready for "Second-Pass Filecutting". Basically, this is doing a second pass on cards to review the work of another filecutter. If you are interested, you can check out the guide for this here: Second-Pass Guide. Once you've reviewed the guide, get in touch with a lead to get a second-pass assignment.