Jump to content

Voice-Actor Application - Miles B. Huff (Sweyn78)


Recommended Posts

I'm around, but haven't heard anything more from Miles than what he last posted on this thread. I don't know if he's still available or not, nor do I have another way to contact him besides on this thread.

Sweyn and I have been in regular contact over email, he's going to work on some voice recordings for us in the next few weeks. We've also now got female voices in 0 A.D., and another female voice artist is being recorded currently to add more variation. We were also contacted a few weeks ago by another male voice artist who is also now recording some phrases for us, so his voice will be used alongside Sweyn's voice for the male units.

These are the final phrases that were recorded: http://trac.wildfiregames.com/wiki/Audio_Voice_List

Link to comment
Share on other sites

  • 5 months later...

Alright, so my Windows computer's PSU went bad and the rest of it is about to (it's mobo is from 2004, it has a P4, etc). My mic didn't work on my Linux computer (this laptop) over the summer, and I didn't have access to a Windows computer at that time (my roommate's was fingerprint-protected and he was never here to unlock it for me, and the computer labs required silence). In two weeks, I will be home for winter break and will have access to a computer running Windows in a very nice recording environment.

We may not need to wait until then though, because I think I might be able to get it to work now on my laptop. I'm going to give it a go and, if it works, start recording immedietely.

Ugh, I feel terrible for having not done anything over the course of a year. I'm sorry. :(

EDIT: Yes! I got it working! :D

I'm recording now. :)

Edited by Sweyn78
Link to comment
Share on other sites

Alright, I've just recorded the following:

  • What is it? — τί εστι; — tí esti? — [tí esti]
  • My lord? — δέσποτά μου — déspotá mou — [déspotá muː]
  • I will gather together — συλλέξομαι — sylléksomai — [sylːéksomai̯]
  • I will herd — νέμω — némō — [némoː]
  • I will go out against — ἀντέξειμι — antékseimi — [antéksiːmi]
  • I will attack! — eisbàlomen
  • I will repair — siachno
  • I will hunt — kynegetèo
  • I will heal — iàomai
  • I will march! — porèuomai
  • I will retreat! — hypochorèsis
  • Battle cry — alalài, alalalài!
  • I will garrison — parafulakh

I still have these to go:

  • I will walk — βήσομαι — bḗsomai — [běːsomai̯]
  • I will build — τεύξω — teúksō — [teǔ̯ksoː]
  • I will work land — γεωργήσω — geōrgḗsō — [geoːrgěːsoː]
  • I will fish — ἁλιεύσω — halieúsō — [halieǔ̯soː]

I have not yet edited anything.

Link to comment
Share on other sites

I've updated several of the files, as well as recorded a female version of Ti Esti. I'm getting fairly good at doing a female voice, so I can voice everything in both sexes, if you want.

Same link as before (it automatically updates): http://dl.dropbox.co...-Sweyn78.tar.xz

I'm experimenting with them in game right now. :)

Note that the female voicing that I can do probably isn't as good as what someone whose voice had never androgenized could do, so I guess any such recordings I make would be just decent solutions until a better female voice can be found.

Edited by Sweyn78
Link to comment
Share on other sites

Oh, maybe I have an older version then. The only female voice in mine is a sighing sound.

EDIT1: Yeah, I'm on Alpha 10. That would be why, lol.

EDIT2: Just added the 0AD ppa from launchpad. The new version is great! The graphics, especially where water is concerned, are outstanding, and the female voice acting is indeed quite nice. It is a bit quiet, though, and I noticed that being the case with my recordings, too. It would be nice if there were a text file or something that contained a list of the different voice recordings and the volumes at which they should be played. For example, the default volume level for speech could be 133%, and shouts for battle could be 150%. You could also maybe add a speech volume slider to the options like many other games do.

EDIT3: Just had a look at the female recordings. They haven't been normalized, which is one of the reasons why they sound so quiet in-game. I would normalize the ones I have, but they're .ogg (a lossy format), and I know you would rather all edits be done to the original uncompressed versions.

Edited by Sweyn78
Link to comment
Share on other sites

Okay, I inserted my voice acting into the game and realized that I need to decrease the pitch a little on most of them; the game automatically changes pitch to keep the sounds from all sounding alike, but since my voice is in the alto range (I made the recordings speaking as low as I could), even raising the pitch a little bit makes the voice acting sound odd.

So, I went ahead and did that. Here are the new files (same link as before): http://dl.dropbox.co...-Sweyn78.tar.xz

Btw, I'm not sure if it already does this, but having the game automatically adjust tempo by a little bit might be another cool effect. :)

EDIT: Ah, I'd misuploaded the new archive when I originally posted this; the right one is up now. If you downloaded the archive before 9:30 UTC-5 29/11/12, you downloaded an old version.

Edited by Sweyn78
Link to comment
Share on other sites

...and the female voice acting is indeed quite nice. It is a bit quiet, though, and I noticed that being the case with my recordings, too.

I believe someone else on the team complained about the new female voices being too loud when I first committed them, so they were quietened in the soundgroup xml file (which also controls the pitch variations you mentioned). There's no need to edit the sound files themselves.

Link to comment
Share on other sites

In my opinion, the audio quality of the female voices is not up to snuff and I am secretly hoping they will redone on better equipment before the release.

I'd be interested to know if others feel the same way. It was generally agreed at the time that the new female voices were much better quality than the male ones we still have on the svn now.

Link to comment
Share on other sites

Female #2 was a noticeably lower quality than female #1. Also, some of the pronunciations were imperfect, but that's the case with mine, too (for example, my dialect of English diphthongizes [e] into [ᴈi], so I realized /e/ as [ᴈi]), though my errors were much subtler (the female speakers, for example, tended to reduce /e/ and /i/ to [ᴈ] and [ɪ] in unstressed positions, which is a feature of English, not Ancient Greek; I did not do this). The female voices also diphthongized and centralized /o/ (I made sure not to do this, but my pronunciation, although closer, was still incorrect; I realized /o/ and /ᴐ/ as a [ɒ]; my dialect of English doesn't have [o] or [ᴐ], and I'm not very comfortable with pronouncing either of them because [ɒ], which I do have, is so similar, meaning I've never really had to practice them in order to speak other languages well). They also used [æ] in some places, but it wasn't in Ancient Greek. Female 1 aspirated her plosives, which is a definite no-no when pronouncing Ancient Greek (incidentally, the recording in which this is most apparent, I did a good female recording of, so if aspiration bothers you, you can use mine, lol). Also, in "ti esti", you'll notice that female #1's first recording uses English's question intonation, which is especially bad because the pitch-accent for "ti esti" is high on "ti", but in that recording, she not only didn't put the high pitch on "ti" but she put an unnecessary one on "esTI".

So, yeah. The female recordings sound nice in-game, but they are fairly Anglicized and female #2 needed a better microphone.

Also, what male ones are on the svn?

And regarding the ideal loudness for the voiceacting, it should be whatever the minimum loudness is for it to be A well-understood & B correct in effect (like, "eisbalomen", "alalai", etc. should be louder than most other recordings). With the current female voices, I feel the current volume is generally good for B, but is lacking for A.

Also, ((I might have already mentioned this)), the pitch variations are currently a bit too wide, occasionally making the voices sound weird. This doesn't affect the female voices as badly (they don't have the rougher vibration of the male voices), so maybe you could reduce the female pitch fluctuation by a little and the male fluctuation by a lot; that way, no voices will be odd sounding. Generally, you don't want to adjust pitch by too much more than 6% in either direction (many things can be, but this is usually a safe value that prevents things from getting messed up due to the DSP), so I'd say do +5% & -3% for male voices that I made (they were already each lowered by about 5.5%), and +/- 6% for female voices. Also, as I previously mentioned, changing the tempo at random would be a neat way to add some realism to the voices, in addition to the pitch alterations. Try to keep the tempo changes within +/- 5% (really, +6% & -4%, because slowing down is harder) or so.

You should also variate amplitude between different files. I'd say no more than a 1db shift in either direction (mine are normalized to -1db, which is the default for Audacity).

If you really wanted to get fancy, after the initial pitch&tempo&amplitude change, you could make a copy of the edited file in memory. This copy's pitch&amplitude could further fluctuate by +/- 1% & 0.25db (not tempo though, due to the way I'm going to suggest the sounds come together). You could then have the original edited sound fade out and the copy fade in, each linearly, and then blend them together. This would not only mean that pitch&tempo&amplitude fluctuate between sounds, but that pitch&amplitude fluctuate within sounds, and that's cool. :D

To further the fluctuation of amplitude within sounds, you could put a weak vibrato effect on each copy. I don't use the vibrato effect in audacity much, so I don't know good and subtle values for this, but randomly fluctuating the strength and period of the vibrato in each soundfile before the merge could really be cool. I'm going to go ahead and say the maximum strength shouldn't be more than +/-0.25db and that the period should be centered on the average spoken length of a vowel and not fluctuate by more than the difference in time between a normal and a stressed vowel. You could also throw a tremolo effect in on top of that for added awesomeness. I'd say the strength of a tremolo shouldn't be more than 0.5%, and that its period should follow the same rules as the vibrato's period.

EDIT: In my typical speech, stressed vowels are about 0.1secs and unstressed about 0.2secs (yeah, odd, right?), so I'd say the ideal period for me is about 0.15secs +/0.05. Remember that this only goes for my everyday speech, which is nearly 200wpm. You will probably need to find different values.

Edited by Sweyn78
Link to comment
Share on other sites

  • 2 months later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Create New...