[OUTDATED] Acted Voiced Dialogue
THIS THREAD IS OUTDATED PLEASE GO HERE:
Current Release: V002,
Voice over main page
This page is for managing a voice overs mod and who is commiting to the dialoge of specific characters. Please click on the link to see the character. Things to keep in mind when deciding which characters to do is: Age, Sex, Wealth (upper class, lower class), and Race. These attributes usually help determine how someone sounds in Elder Scrolls. Please send me a message to let us know who you would like to do.
Recording Tips and Tutorials:
TES Alliance Workshops - Guides you through the process of creating voice-overs for Oblivion, using Audacity, a free audio editing program
- TES Alliance: Workshop 1 - Introduction to Audacity
- TES Alliance: Workshop 2 - Working with Audacity
- TES Alliance: Workshop 3 - Creating .LIP Files
- TES Alliance: Workshop 4 - Mic's, Signal-to-Noise, Recording, and more
- TES Alliance: Workshop 5 - Acting Guide
- Video on how to produce large amounts of .LIP files with CS v1.0
When recording please use a pop filter or at least ladies nylon (put the mic into the nylon) this will help get rid of bad "s" and "p" sounds that plague voice overs. Speak clearly. Use a heavy blanket to put behind the mic. This captures reflections of the sound from coming in through the back fo the mic. This will allow a cleaner voice.
Please use compressors on your voice. You want the compressor settings to be: Attack (fastest possible) Release (slow) Ratio (high 10:1 to even 20:1) And the Input (threshold) and Output (gain) You want to set it to a setting that doesn't make the volume shoot up and then go really quiet.
- Video tutorial on compressors
- Video tutorial on voice overs (but he doesn't do an example)
- Video tutorial on voice overs and compression
- Video tutorial about doing hip hop vocals, but it has decent info on what you can do with voice.
Audacity video tutorial
This tutorial seems good, the guy know sound manipulation and explains things simple. I am digging this guys stuff. This tutorial gives the basics, but this guy has others that go futher on sound manipulation. You can actually sound like a completley different person, after you have experience, but for this project we need it to "clean" the voice file and just keep the voice tones together.
- Audacity, a free audio editing program
- Switch Sound File Converter, to convert WAV to MP3
- EggTranslator v22.214.171.124 (Will export all individual NPC dialogue to the HD, around 961 files.) Try to use v126.96.36.199, that one works for me, v188.8.131.52 is bugged
- Construction Set (both versions, the Construction Set Extender, and Oblivion Script Extender
This is a recording interface at guitar center for like $50, that are USB, come with a program, and are more professional than pluging a mic into your sound card. Instantly less noise. Sound card noise is the #1 thing I've heard so far. It is also the hardest to remove. If you can't afford an interface, don't worry. Getting in character and getting it recorded is more important then waiting. This is an alternative, better interface but it is $85.
What I use is going on ebay for $150, but I think it will only run on XP or mac. I like what I got - it is a PCI card, so it is right on the motherboard, so it has great latency, I can record on the lowest buffer size and hear my voice with no delay. It also has some great preamps, It has the high end focusrite preamps. This used to sell for $1500 in the 90's
Please keep the original wave format and mp3s in Oblivion are sampled at 44.1 Hz frequency, 64 Kbps bitrate and Mono channels. See the TES Alliance Workshops above for more information
Right now I wouldn't worry about "voice talent" If it really doesn't work, it doesn't work and I hope you understand. Personally I think everyone can do a character. And if you do one character, and it doesn't fit, then the solution is that your voice is built for a different character. Sound is relative like that. What doesn't sound good in one situation sounds brilliant in another.
- Go into the CS and find the person's dialogue and the corresponding audio file. Look in dialogue filter under Character.
- In the filter box, look for 0(then the persons name) example. 0caiusScosades
- In the editorid box, click on the top one.
- To the right of that box, you see another box with the headers: Quest, Info, NPC's,... etc..
- Under NPC's should be your characters name. If you don't see your characters name (the filters not perfect). Then you just click on the second line in the editorid box. Keep clicking to the one below untill you find your NPC's name under the NPC header in the box to the right.
- Next click the line with your NPC's name.
- Then look down at the box below the one you are woking on. This the the responce text. This is where the dialogue is. Double click a line of dialogue.
- In Responce Text, is the complete text of the dialogue.
- In Voice Filename, that is the name your mp3 will have to be to sync up with the game engine. You can not change this you have to name your voice file of that specific text to that specific filename.
- Record it as an MP3 and a WAV and save as the corresponding file name.
- Under the Voice Filename you'll see a header call Path. This is the path that the file must be in. No execptions. You can not change.
We do need to optimize this process. I am recording the files and will double back over after the mp3 conversion and place them in the right path. I am trying it out for Cosades this way, to see what kind of "product" this way produces. I did it a different way the first time and honestly I like what I am producing in this process.
For MP3 comversions I am using Switch Sound File Converter. I am doing this because it will "normailize" your audio file. Which means it will be at the right volume for the game (in short).
Ice's Noob Tutorial for Recording:
In the pictures I labeled parts of the CS based on what step you should click/look at that part of the CS. So when there's a "3" in the picture, that means look at it as you read step 3.
1) Open the Construction Kit
2) Load Oblivion.esm and Morrowind_ob.esm. You don't need to set an active file, and you don't need to save.
3) In the object window double click Actors>NPC>"You're character". For the tutorial, we'll use Vivec, so double click Actors>NPC>VivecRace>Male>fbmwVivec
4) Click the "Dialogue" button. The CK will load for a bit.
5) A "Dialogue" window will pop up. Notice that in the top left corner it says "Filter." Below "Filter" should be the name of the NPC you're voicing.
6) Notice the list of Topics Next to the filter. They're sorted by editor ID. Click on an editor ID to see the topic info. Look under the "NPC" Column for your selected NPC. If the NPC column is blank or has an NPC we're not voicing, click the next Topic ID. In our example I found "1RedSMountain" and that topic had "fbmwVivec" in the NPC Column. That means that Vivec speaks this topic, and so we need to record dialogue for it.
7) Double click on the topic info (see picture). That brings up the "Edit Response" window. This window shows us the response text (what Vivec says), his emotion, and the voice filename/filepath (highlighted). Sidenote: If you wish to include an esp file along with your sound files, then consider editing the emotion type and value while you're here.
Use Audacity to record you as you read the response text. If you need to know more about Audacity / recording, look into one of Sot's tutorial links, or browse the thread. Basically you need to record a line, clean it up (use a compressor, use noise removal, etc) and save it as a wav.
9) When the file is ready to go, convert it to mp3. Please keep the original wave format; mp3s in Oblivion are sampled at 44.1 Hz frequency, 64 Kbps bitrate and Mono channels. See the TES Alliance Workshops above for more information. (This is easiest to do with Switch sound converter. Find the link above).
10) In the "Edit Response" window there is a box called "Voice Filename" (#7). Below the voice filename is a box that contains the NPC and a column called "Path." For Oblivion to play the file, the mp3 must be in the right format (like we said in #9), and it must have the EXACT filename as the text in the "Voice Filename" box. This is easy to do because you can double click in the box to select the filename. Copy the value (Ctrl + C) and then right click your voice file>rename and paste the new filename in (Ctrl + V). Finally, place you're properly formatted and titled mp3 into the correct filepath, creating folder's where necessary.
11) Test your voice files in game! If they work and you like them, create a zip that contains your files placed in the right folder path. (Basically, somewhere on your hard drive start creating new folders: create a new folder and call it "Data," inside that another new folder "Sound," etc etc. Then post a link to them on this thread!
12) Thank you!
- Sellus Gravius Completed: iceburg333
- Bacola Closcius Completed: iceburg333
- Caius Cosades Completed: Sot
- Hasphat Antabolis Completed: iceburg333
- Sharn gra-Muzgob
- Duvianus Platorius assigned: leftpassenger
- Mehra Milo
- Hassour Zainsubani assigned: robertneville777
- Codus Callonus completed: iceburg333
- Nibani Maesa
- Raesa Pullia
- Dagoth Gares assigned: Darth South
- Divayth Fyr assigned: iceburg333
- Yagrum Bagarn
- Mehra Milo
- Gilvas Barelo assigned: iceburg333
- Vevrana Aryon
- Athyn Sarethi
- Miner Arobar
- Garisa Llethri
- Hlaren Ramoran
- Brara Morvayn
- Bolvyn Venim
- Crassius Curio assigned: leftpassenger
- Yngling Half-Troll Completed: iceburg333
- Dram Bero
- Orvas Dren
- Velanda Omani
- Nevena Ules
- Master Aryon
- Mistress Dratha
- Mistress Therana
- Master Neloth Completed: iceburg333
- Master Baladas
- Sinnammu Mirpal
- Sonummu Zabamat
- Savile Imayn
- Tholer Saryoni
- Danso Indules
- Vivec Assigned: soucy666
- Argonian Assigned: Solin
- Bosmer Assigned: iceburg333
- Imperial Assigned: Sot
- Fargoth Completed Kefka
I wouldn't mind helping out. Provided my voice would suit the charachters?
I'd be up for a few voices.
You have a seemingly good plan Sot, but it doesn't quite work out that way.
If we're talking about general greetings, you don't neccessarily want go person by person, because many people share the same greeting dialogue.
For example, a Redguard in Balmora might have a specific greeting. If you record this greeting, all Redguards in Balmora will use that the same audio for that greeting. So it's kind of a good thing, because you're essentially getting more than on person done, but you must be very careful what topic and race is already recorded.
Now if we're talking quests (like specific missions), you are correct. For the most part, quest dialogue is unique to every person, so you could go person by person if you want to record for quests.
Anyways, this is pretty much what you have to do, and what makes this process so toilsome and a pain in the butt:
- Go into the CS and find the person's dialogue and the corresponding audio file.
- Record it as an MP3 and a WAV and save as the corresponding file name.
- Make a LIP file in the old CS.
- Rename the LIP with the same name as the corresponding MP3.
- Put the MP3 and corresponding LIP file into the correct folder and you should be good.
And this is just for one audio file. It gets pretty confusing when you have to juggle multiple audio file and LIP names, and making sure not to mix up the names.
So...Sot good buddy...I think you're a prime candidate. I mean, you're the man with a plan and all
I would do this, but I'm currently working on getting Morrowind Conversation into Oblivion/Morroblivion.
But if you need help, I'm always here.
I still say screw the lip files for now, just have generic mouth movements anything to speed things up.
I say, if a person is doing voices, he must finish the same race. If a dunmer has three-or four voices(they already have two right now), it will be illogical. However making voices for Morroblivion will be the ultimate victory of Morroblivion, sealing its fate of being better than Oblivion(which I think has already happened with Vis. Overhaul.
I think it would be wierd playing, and hearing my own voice come out of other players . But I would definitely do it. I've no experience with the CS really, so, all I could do is record, and upload the voices and let someone else organise them. I'd need the scripts too obviously XD
Its won't be that odd, because you won't be doing your voice, unless your natural voice is very similar to one of MW's actors. Which voices can people do? (Remember its MW voice actors you trying to imitate not Oblivions, some are different).
We are talking full dialog here right? Not just general greetings? I'm not sure what I could do, I guess the best way is to have some kind of audtions for parts?
Actually recording voices wouldn't be such a hard process if you used the Construction Set Extender (http://www.tesnexus.com/downloads/file.php?id=36370) It fixes the Generate Lip File button so that you don't have to use the old CS.
I was actually looking for a canidate to help me manage the voice overs. I can do voices, but I don't understand the CS's dialoge to get it done. I was going to start with the main quest and just do it person by person. Here later tonight I will update this main post with characters and will be starting to put people down as they commit to giving us the voices for those character's dialoge.
After the main quest, we should do guild by guild, house by house, then city by city....
As far as greetings go, I say lets not worry about them. Let's get unique voices down first. I for one would love to push for an 11.11.11 morroblivion release date (with voice overs) and don't mind adding greetings after that fact (if we don't have time)
Nice, thank you for sharing that! That will make my life a billion times easier.
Hey Sot, I will assist you the best I can. Just send me a PM or post here on what you need help on.
Always a pleasure to help.
I have updated the main post
Dagoth Ur is already done ... and we have the orignal videos (and hence voice) for Azura as well.
Hashmi I think you should be Vivec, if you have time.
updated, I put you down for hassour zainsubani.
I picked Caius Cosades
Hashmi I think you should be Vivec, if you have time.
Now, why do you think that?
I am not a native speaker (am from Pakistan) and I don't sound anything like a Chimer ...
You used to have Vivec in your avatar, and also you're like a god in this project. As Dagoth UR already been done, you definitely deserve to be Vivec.
Hashmi, not being a native speaker doesn't matter... sound is relative. There is always a beatuful way to use any sound/voice. You might find that Vivec might suit you perfect. Since he is (in essence) his own race, your accent (if any) might work.
I am ONLY thinking about Dagoth Gares, how do we do that? How can I find Gares' dialogue to read?
Well actually, Vivec does have voiced dialogue, albeit like one or two audio files. But at least this will give you an idea of how Vivec should sound like.
By the way I have recorded all the dialogue of Dagoth Gares. Should I send you these voice files, or try to make them on my own.
By the way recording is fun. I can also do some others, but linking might be a little tough.
both ice eyes and robert have volunteered to add in the dialoge and create lip sync.
i would zip up the mp3's and post a link so one of them can put it in.
what we need to do is deside how best to merge these, and come up with a naming system. Should we label each .esp as they name of the person, or come up with a mainquest.esp?
we should ask david what he thinks...
updated the main post
Here is the link to my dialogue: http://www.mediafire.com/?qqrhkdnv7knvbdd
I've named the mp3 files with the starting of their speeches.
Also I've put them in folders according to topics
There is an extra wav(Yes, the format is wav,I haven't converted anything. Hope it won't bother you.) which I can't remember where it belonged.
If there also is a missing dialogue, tell me and I'll send you.
I've also started Yagrum Bagarn!
I would love to help I would especially like to do some sort of Daedra/ God
this is a quick one I did for Vivec.
I'm happy to do anything you want me to just ask :)
I'm very happy to see that we are finally to the point that voice overs are being done. I have nothing to add to this other than that a voice mod for vivec has already been done. I haven't looked into how it sounds exactly, but if time could be saved by incorporating it and it is of good quality, why not use it?
I look forward to hearing all of the voiced characters.
Edit: Just posing a potential problem, can the vivec creature be properly lipsynched without any issues?
Cool I just got home from work and I am checking these files out. Thank you!!
I don't know if we could use the files from someone elses mod. As far as voice sycning him. I say he talks with his mind...
What happen to doing all the voices by race?
@Darth South. You have quite a bit of fuzz on your recordings. You should use a noise removal tool to fix that. Right now that will be very noticeable in game.
@Si9mth, that recording is a great example of what happens when your not writing with voiced dialog in mind. A lot of MW writing does transfer smoothly to spoken dialog. I am not suggesting altering the original script should be done, it was just a thought on my mind.
I think the voice files should be named according to what the CS names it. That way we don't have to hint down each file and rename it. That will be a pain.
Might as well do this because you have to go in the CS to determine what emotion the voice and LIP file is.
Rob has the right idea, the CS naming convention is a more efficient and consistent way to do things.Also I though OB used 64k mono mp3(something like that), can you go higher? Is that even a good idea(extra space for no reason)?
I had just posted a thread on voice acting and David Budreck was nice enough to redirect me here. If you still need any help on voices, I would be glad to help. I don't know what race/character my voice would be suitable for, but if it comes down to it, I can make a sample recording for you. Either way, I love the idea of getting voices for all races and characters whether it be generic conversation or quest related. I messed around with making my own character and custom voices a while ago for Vanilla Oblivion, but v1.2 of CS doesn't allow me to make the lip files.
I hope all goes well, this sounds like a great idea.
This thread has not been updated in seven weeks.
This project is dead unless you will take the lead and try to see it through.
That's a bit of a mood-killer, but if I find I can get the time, I would love to take it up. My fiance is a fan as well, and he would lend his voice as well. I'll have to get in contact with a few people I know who are in to the sort of thing, but in the meantime, I'll just watch the forums and get a better idea how this place is and such.
Morroblivion is a VERY Small, nich site. It is banned, so technically we are underground.
People keep expecting this site to be like Nexus, but take a look at the number of Registered Users logged in at any one time. Not many. We definitely need people to take the lead rather than waiting for others to do it. Otherwise, as with this thread, it will probably be a very, very long wait.
If you want something done right...
That completely makes sense. I only just found this site today. I can understand why it's underground. But "if you want something done right..." is something I agree with. If I can figure out how to attach the voice files to the appropriate spots, then I can take this up. I am not very good at using the CS yet, but I am sure I can find tutorials for it.
Oblivion wants voice files to be named in a certain way, and placed in certain folders.
If you go to a line of dialog in the CS and double-click to open it. The window will tell you want it wants the voice file to be named, and where it needs to be located.
There are no options for naming/placing voice files where you want.
Any help is needed for this. It's pretty much, record what you can and post the files up here.
I'll update the main post with pointers that David has given us.
I've personally am going through Cosades for the second time. I had it done but my comp went through a strange mini crash that erased some things....
I am hoping to have it posted by the end of this weekend.
Mehra>> If you post some voice files I will help narrow down some persons for you to try out, the main thing is a descent mic and some sort of baffling to go behind the mic. You can use a heavey blanket if that is what you have. Also a descent mic stand helps out alot. If you don't have a pop filter to go around the mic you can always use laddy panty hose or a sock. That will keep the P's and B's you say, keeped from jumping up in volume.
Sure, I'll upload a sample or two, with generic dialogue and if I fit somewhere let me know. I have a few people who wouldn't lending a hand if you need extras. And in response to David, I was aware of the sound names, just not sure if I plug them in right, then again, I was a dork and didn't save my test file back when I first tried to make a custom character. So that was on me.
the main post has been edited with better instructions.
I can to a very simular voice of the Dunmer. Dunmer male that is. Do you still need it?
I think I could help out with the some of the raspy ashlanders (singing in a metalband and blazin like its 420), Argonians and Khajiits, I'm pretty good with accents and I think the Khajiit and Argonian accent should'nt be hard and can even be improved upon.
Ofcourse Argonians and Khajiit hardly feature in the main quest, altough I can do Skink from Sadrith Mages regardless, I think I also might be able to do a pretty good Diyath Fyr, highborn, intelectual with just a dash of madness and some other Telvanni perhaps.
I just saw this topic for the first time today, and got very excited. I've emailed about six friends asking for their help / their voices, and read through this topic. I think I get the gist of this and I'd like to try to add some voice (as well as try to convince my friends to do the same. One already has expressed interest). I'm a 20 year old male, 185 lbs. I don't think I've got the rasp of a dunmar or beast, but I think I could do nords and orks.
I'd like to try out for Yngling Half-Troll. I'll record some experimental sounds and upload them.
In the meantime I had a couple of questions:
Is there a good way to access what spoken dialogue a character does have (For example, is there a way I could listen to Yngling Half-Troll outside the game to get a feel for his voice?) I looked in the editor and found where voice files could go, and found some files (those under Oblivion\Data\Sound\voice\Oblivion.esm\Nord\M) (or Oblivion\Data\Sound\voice\Morrowind_ob.esm) . I was thinking that there are more voice files in my morrowind data files: Morrowind\Data Files\Sound\Vo\n\m. I wanted t osee specific lines that Half-Troll had, but didn't know how to locate them, so I guess I'll just listen to generic Nord to get the idea. I guess what I'm trying to say is it would be nice to have a way to access quickly all the files said by a certain character, so that I know I need to record a file and that it hasn't been done before, and if it is already done, I can listen to it... But this may be impossible.
Is there any way we can export scripts for people who don't have access to the game / Tes Construction set? A couple of my friends don't have the game, but would be good voice talent.
How do we give you the sound files? I assume we upload them to a site like media fire and then post a link on this topic. Since I havn't seen that, does that mean people are PMing you their links?
I assume that if a character already has a voice file for a specific line, we aren’t replacing it. (I guess this is a 'duh' moment, oh well).
If we do a character (say Half-Troll) Are we for now doing just the main quest dialogue or all of their dialogue? I assume you'd want all of their (Specific) dialogue so that you have character consistency.
If the character doesn't have a voice file, are we adding one, even if it’s a generic line (like not one that's specific to that character / that all races say). It seems to me that we should have a person per race / gender doing all of the general lines. Then we should have other people doing the specific characters’ specific lines separately. This would lighten the load of the majority of volunteers. Otherwise each volunteer is voicing a huge amount of general dialogue that really only needs to be done once for each race/ gender.
We could do this by committing a person to each race/ gender (these would be huge roles). Then other people could look up their characters (say Half-Troll, just for the heck of it), and see what quests they are in using the wiki. I think then that they could use the filtered quest dialogue to do only those lines that the npc needs for their quests. For example:
Half Troll’s first quest is Hlaalu Hortator. In the editor that’s: fbmwMQHlaaluYngling. Then I see only his topics for that quest. Presto. I record just those files. I don’t have to record (along with everyone else) the generic files. While this would be a slight (and only slight since the generic voice would provide consistency) hit to realistic-ness, it would save not only hard drive space (less custom voice files), but also time and effort, which are huge commodities on an ‘underground’ site.
I am going to start with Half-Troll, doing only the specific files. If this goes well, and if I have the time and desire (this is a big if: no promises) then perhaps I’ll volunteer to do all male Nord generic lines. (That is, if this ‘general lines-specific lines’ grouping method works for people.)
Also, if this method is too much weight on the people doing whole races, then we could break that up into two or three people per race, which really would be fine, I’d think. We would just need to make sure we had good communication.
At this point, it would be impractical to do unique voices for all 1300 plus NPCs in the game. However, if we use the General-Specific method, then a voiced Morroblivion would be achievable, and we could always go back later and add more custom files in updates.
Wow. That was a big post. But I’m excited. I’ll get to work on Half Troll, and try to rope some people in on this. Please let me know what you think!
PS. Where can we go (or what can we listen to) for a prenunciation guide for words like Hlaalu, etc?
I prounounce Hlaalu Huh-la-loo(I don't prouncoune it exactly like that but that is the only way I can textually say how I prounounce it).
Here's a test of me doing Half Troll. I really don't have the depth and grit of a nord (which rules out orcs and beast races for sure), but I can do that bit of Scottish accent they have. If you're fine with these samples, I'd love to continue with Half-Troll.
I could also use audacity to lower my pitch, that seems to sound alright, though I don't know how I could add grit.
(ps. I'm also definatly interested in character suggestions as well. My brother may also be interested, his voice is slightly higher and has a higher range I think. I'm going to try to get hime to do Neloth.)
Vivec: Thanks man. I've always said Hi-la-loo, and that sounds pretty close, so since its muscle memory and close I think I'll go with it. Thanks man.
EDIT: I am also wondering:
Can we do a massive generate LIP Files, or are we going to need to generate lip files for each voice file, and then make esps that need to be merged?
I've finished Half-Troll's main quest lines. (Not his side quests yet) I put the voice files into the appropriate folder location (tes construction set confirmed their presence, saying the lines had the matching files. The files would play in the set). However, when I went to half Troll in game, the lines (I only have one since my character has finished the main quest) appear silently. At first they rushed by, but after I corrected the location, they pause as if its waiting for the wav file. But there is no sound. Did I miss something? When I try to generate .lip (I'm using the CSE that was suggested in this topic) it deletes my wav files! Mp3s don't seem to work either. I've linked the main quest mp3s that were converted from the wav files using the suggested program: http://www.mediafire.com/?614t9feh84z2aao
iceburg, i am going over your post. I'll hit up in a bit. But I do agree, we need to have one person do the generic gender\race dialogue, and then use specific people for specific NPC's, I thought about the main quest first, so that if this looses steam, we at least have the main quest done....
This progect is huge.... but I do almost have Cosades done. Then I wanted to do generic Imperials. That way Cosades will be a fully functional voiced NPC.