I’ve finally decided to take the plunge and start recording gameplay in FFXI for YouTube with my avatar sitting in for me.

Last Sunday I recorded a little over 3 hours of gameplay and was quickly overwhelmed with the editing process. The events of last Monday and the remainder of the week put this project on hold.

I decided that for this project to go back to the model’s original hair and color edit it. I still plan to make custom hair for it in Blender, but I’ve been running into problems with it. Mainly I’m finding I have to relearn everything after taking a break from it that extended beyond NaNoWriMo. Changing the hair meant removing the glasses since the ear pieces proved to be problematic once the hair was no longer hiding them.

I am using the Natural Reader Commercial program for the avatar’s voice. Specifically, I’m using the child David voice in the US English list. The first thing I’ll say is that I acknowledge that I have tone/pitch loss so take this with a grain of salt when I say of the voices in the list, this one felt less robotic/tinny to me. The rhythm and cadence of this voice is pleasing to me and feels like it’s a good fit for the Toadie persona. I just need to remember to not let it sound breathless, even though I’m guilty of that in person.

Originally I had recorded my voice using Audacity while playing the game but since my mic is connected to my headphones, I quickly discovered that thanks to my hearing impairment, the game’s sound was picked up as well. That and speaking on the fly with no script or notes is not a strength of mine. Too much rambling – more so than what I’m guilty of in my posts here.

Seriously, I’m not used to verbally streaming while playing a game, especially if I’m by myself, so I have no idea what I should be talking about. However, I find it easier to watch the clips and write down commentary about what I’m seeing. The reader program can then convert it to speech. This should help keep the dialogue relevant to what’s going on in the video.

The bonus in this is that the voice from the reader comes free of noise so using the Face and Lips program to animate the face of the model is much easier. The only sound waves to look at are from the voice. So it’s just a matter of deciding which mouth shapes I need to match up with what I’m hearing. Having knowledge of the phonetic alphabet and being a lip reader makes this part a piece of cake for me. It’s still time consuming though.

Posing the model frame by frame was still done with the MMM program and rendered separately. For whatever reason, Filmora doesn’t have the codecs to use the video straight from there. I had to use Windows’ Movie Maker to convert the format. I needed Filmora to get the picture in picture overlay to happen. For whatever reason, the plugin that does it in Movie Maker isn’t working and I think it’s because I’ve got Windows 10, which technically doesn’t support Movie Maker.

I plan to create roughly 15 minute segments for this video series, so it’s a matter of figuring out what’s important and interesting in all of what I’ve recorded so far. Right now all I have finished is the introduction of the first episode. I’ve uploaded it onto YouTube as an unlisted video so I could share it as a sneak peek with my readers. I’m not sending this out on my other feeds since it’s a work in progress.

Eventually I’d like to see the avatar have a legit background to sit in rather than just a black backdrop. The chair used here is from vasilnatalie/bandages, which you can download here. I like the chair because it reminds me of the big comfy (but ugly) chair I once had. I have a texture in mind to make it look more like that old chair.