Making A Custom Voiceset

The first thing you’re going to have to do is download some stuff, and as with anything new there is a little of a learning curve. Just be patient and use copious amounts of google and you should do just fine. Besides, it’s all super easy!

Things you will need:
- An SSF Editor
- A TLK Editor
- A USB Microphone
- Sound Editing Software

SSF Editor – You’re going to need this in order to edit your SSF files. SSFEditor, the one I use, I got off the vault and its super intuitive.

TKL Editor – You’re not actually going to be editing the TLK file. You need this editor so you can browse through the TLK file in order to name your voice sets. I use Aussie NWN2 TLK Editor. There is also TabularFramework (2DA-Editor & TLK-Editor) which edits your 2da file as well, but it loads much more slowly than the Aussie one and you don’t need an editor for your 2da file anyway. Any text editor (such as notepad) will do the trick.

USB Microphone – Okay, here’s where the money might start coming in. If you want to do any recordings of any sort of quality you will need some good equipment. Now, I use an icicle brand preamp and an audio-technica condenser mic. The setup cost me about 100 bucks in all. Now, for something as simple as this you’re not going to want to waste a Franklin. A USB mic is going to be your best friend. You can snag one of those for 20 bucks. (Little tip for gamers, if you play Rockband or any other game that has singing then your game mic might plug right into your computer.) A USB mic isn’t the way to go for professionals, but for a hobbyist they’re just great. Why USB you ask? Well a regular mic will plug into your soundcard to record, and if your soundcard isn’t made specifically for recording (it most likely isn’t) then you’re going to have electronic back feed from all the stuff buzzing around in your computer. In other words there will be static. Now, you can edit out this static, but the quality will be that much worse for it. It’s better to plop down the 20 for a USB mic.

Sound Editing Software – Here’s another place you might have to spend money on. If you’re a professional you’re going to want something like Cakewalk or Adobe Audition. Those programs are powerful, but expensive and (in audition’s case) take up a lot of space to run and if you don’t have a good soundcard they won’t be that useful for you. If you’re using a mac then you’re already setup with Garage Band and wont need any editing software. However, chances are you have a PC like myself and need to get your own. For a hobbyist Audacity is going to be your new best friend. It’s small, it’s simple, and it gets the job done. Hell, I have audition and I still use Audacity for the actual recording as it’s less likely to lag on me.


Well, now that you have your stuff you’re going to need to know how to use it. Here’s where the fun part comes in.

First step is recording your set. You’re going to need to make them wav files for this. Now, if you have a professional sound editor then you already know how to use it. So I’m going to focus this tutorial on using Audacity. (Sorry mac users, I’ve never even touched Garage Band.)

These here are the lines you’re going to need:
• Attack - Command Issued - PC/Henchman/NPC
• Battlecry 1 - Yelled while running to battle - PC/Henchman/NPC
• Battlecry 2 - PC/Henchman/NPC
• Battlecry 3 - PC/Henchman/NPC
• Heal me - Command Issued - PC/Henchman/NPC
• Help - Command Issued - PC/Henchman/NPC
• Enemies - Enemies have been Sighted - PC
• Flee - Command Issued - PC/Henchman
• Taunt - A skill that tries to make the enemy only attack you - PC
• Guard Me - Command Issued - PC
• Hold Position - Command Issued - PC
• Attack Grunt 1 - PC/Henchman/NPC
• Attack Grunt 2 - PC/Henchman/NPC
• Attack Grunt 3 - PC/Henchman/NPC
• Pain Grunt 1 - PC/Henchman/NPC
• Pain Grunt 2 - PC/Henchman/NPC
• Pain Grunt 3 - PC/Henchman/NPC
• Near Death - Announcement that you are near death - PC
• Death - PC/Henchman/NPC
• Spell Failed - Announcement that a spell has failed - PC/Henchman
• Weapon Ineffective - Announcement that you weapon does no damage - PC/Henchman
• Follow Me - Command Issued - PC
• Look Here - Announcement that you have spotted something - PC
• Group Party - Command Issued - PC
• Move Over - Command Issued - PC
• Pick Lock - Command Issued - PC
• Search - Command Issued - PC
• Go Stealthy - Command Issued - PC
• Can Do - PC/Henchman
• Cannot Do - PC/Henchman
• Task Complete - PC/Henchman
• Encumbered - Announcement that you have too much in your inventory and are now slowed - PC/Henchman
• Selected - said when you are clicked - PC/Henchman
• Hello - PC/Henchman/NPC
• Yes - PC/Henchman/NPC
• No - PC/Henchman/NPC
• Stop - Command Issued - PC
• Rest - Command Issued - PC
• Bored - Said when you have been standing too long - PC
• Goodbye - PC/Henchman/NPC
• Poisoned - PC/Henchman/NPC
• Thank You - PC
• Laugh - PC/Henchman/NPC
• Curse - Said when something is failed - PC
• Cheer - PC/Henchman/NPC
• Something to Say - PC/Henchman/NPC
• Good Idea - PC/Henchman/NPC
• Bad Idea - PC/Henchman/NPC
• Threaten - PC/Henchman/NPC

You can write down exactly what you want to say, paste that up there into a text file to keep open and just improve (which is what I do) or even snag voice clips from other games or movies and put it in their place. Whatever your method you must have all those lines conveyed in some way.

Now, using Audacity to do this is like pie. You open it up and click the record button (that is the red dot) then record. That’s it. No formatting or fancy things. When you’re done give what you recording a listen. You might have blown into the mic by accident or maybe you just don’t like how it sounds. Hit the little X button on the corner of the recording and do it again. (Little tip on mic blowing, keep the mic at the corner of your face instead of in front. That way breathing and blowing sound more natural and not like you blew into a microphone.) If you’re happy with what you did (be sure to cut out the silent bits at the beginning and end of the file) then highlight the sound, go to Effect-Amplify. (For audition users this would be called normalize.) You’re doing this so you can hear the voice at an appropriate level in the game. Listen to it again to see if you missed any inconsistencies. [Little tip for people using a regular mic instead of USB. (Even though I told you not to.) You’re going to want to wait and record some silence first. This is so you can edit the static out later. Highlight the silent staticy bit and Effect-Noise Removal. Click the Get Noise Profile button. After that highlight the whole recording and click Remove Noise. Cake.] Play around with the effects a bit (if you’re happy with what you have and are afraid of losing it then just save first) sometimes you can discover really neat tricks that way.

When you’re ready to save File-Export as Wave (or save as wav) and name it something like voice_attack. If you plan on doing more than one of these uniformity is key. Always naming things the same way but changing one part (like what was in front of that underscore) ensures that you don’t make silly typos as well as saves you mounds of time. Now me, I name my stuff exactly like it is on the list. voice_attack voice_battlecry1 voice_battlecry2 etcetera. When I make another soundest I just change the first part. anothervoice_attack anothervoice_battlecry1. It doesn’t matter how long the file name is or what you name it as long as it is a WAV file. It can’t be anything else or the game won’t recognize it right.


Once you have your recordings done the rest of the steps can be taken in any order you want. I don’t do them in a specific way though it helps to if you want to make sure you don’t skip something.

So let’s edit our SSF then.

You know that editor I had your download? Well open the sucker up. Now, there’s a little catch here. If you’re making one from scratch you wont be able to select anything below Group Party. This does not make for happy fun times. You can, however, open up an already made SSF file. Just download one of mine from the vault. (As an added bonus if you named your stuff like I do then you have even LESS work ahead of you this way.) I didn’t really get my format down until the Still Mind set. Extract the SSF file (in the still mind set it’d be named vs_histhrall.ssf) it honestly doesn’t matter what you name this file. Keeping it short of logical will be helpful though. This part should be pretty self explanatory, but I’m going to explain anyway. You’ll notice that you have two options here. Resource ref and String ref. Resource ref is the name of the wav file that you want the game to associate with that action. The String ref is the line of text that might pop up over their head when the talk. We’re not going to touch the String ref. Keeping it at -1 means that nothing will pop over the character’s head. Don’t go typing in the dialog here either. The string ref is a number associated with a line that you’ll find in the tlk file. You’re already going to have to drudge through that to find a name for your soundset. No point in doing it for all the lines too.


Speaking of name, you’re going to want to find that. This here is the fun part. Let’s open up the TLK editor you downloaded and open up the dialog.tlk file. (Some editors already know where this is, but if yours doesn’t you can find it in the default directory C:\Program Files\Atari\Neverwinter Nights 2.) DO NOT EDIT THIS FILE! Seriously, if you don’t know what you’re doing you can really mess up your game! We’re only doing this so we can sift through strings. Either use the find option or scroll through the file to find one you want. The string of numbers is important here. You need to copy or write them down so you don’t forget it. Also, be careful if you’re using the find option in the Aussie tlk editor. Sometimes it will start replacing a bit of word while you’re still typing in what you want to look for. Just make sure you NEVER SAVE when you exit the program in case you didn’t notice it doing this.


Once you have your string (1942 is one of several for Still Mind) you’re going to need to crack open the soundset.2da. (Copy the one from the 2da.zip in C:\Program Files\Atari\Neverwinter Nights 2\Data if you don’t have one in your override already.) You can use notepad or an editor for this. You’re going to see some options here. E. C. Patterson was kind enough to explain these to me when I had that same ‘wtf’ look on my face the first time I opened the soundset.2da.

LABEL = The name of the voiceset that will appear in the TS selection menu and probably the character creation menu as well (for PC voicesets), IF the STRREF is absent (i.e. = to ****)

RESREF = the file name of the ssf, without the extension.

STRREF = the dialog.tlk line number if you want the label to get its string value from that. Enter **** if you just want to stick with the LABEL value and not have to find a suitable STRREF from the default tlk, or create a custom tlk.

GENDER = 0 : Male, 1: Female. Only really important for PC voicesets.

TYPE = possible values : 0-5. Refer to soundsettypes.2da. 0=PC, 1=Companion etc. Type 0s are the most complete (all 50+ voice chats) and it goes down from there. Only type 0s show up during character creation.

In other words:
Folder_That_Holds_Wave_Files SSF_File Number_From_TLK Gender 0
OR
Thrall vs_histhrall 1942 1 0

So pick yourself a blank line (try not to conflict with other voicesets if you can) and put that info in there. Now when you start a new character the voice will show up in the list. Well, so long as you have both the edited 2da and the folder holding your voiceset (with the ssf included inside the folder) in your My Documents\Neverwinter Nights 2\override.


Now that should be it. You’re done. You have a spiffy custom recorded voiceset for your character and can share it with the world if you want.


“BUT SEN!” you say. “I want to make a preview just like you because you are my idol!” Well, strange voice in my head, you can do that by simply looking up how to make a wmv on Google because I am quite done with this tutorial.