Homestar Runner Wiki Forum

in the quote of the week you can always (as far as I've seen) click on "hear it" to listen to a H*R character say the quote. Last time it was strong mad "I don't know when to say when!" this week its strongbad saying... some long line I can't remember.

Anyway, has anyone else noticed the voices in those things always seem more..human? They don't sound quite like strongbad or strong mad, they sound more like a persons immitattion. Do they apply certain sound manipulations to make them sound more unique?

I think it's just you. As far as I know they don't do any processing on the audio. If I were to guess, I'd say it's a psychological: If you can't see the character's mouth moving in sync with the audio, it doesn't sound as "real."

_________________

I've gotta agree with mik on this one, I've noticed that, too, mainly when I've downloaded them. They sound different somehow.

It's due to frequency resampling: the quote of the week is encoded at 44.1 kHz (and I think it's the original take), while cartoons/sbemails are downsampled at 11 kHz which makes you hear 5.5 kHz at most (like through an old telephone), but saves lots of bandwidth

To summarize: for quotes is quality over quantity ('cause they're short), for sbemails/toons it's the opposite.

I'm not sure, man. I just tried recording it, increasing the speed and saving. I then reopened and slowed it down, decreasing the sound quality. Still doesn't sound like him.

I don't know, I don't know much about audio stuff..

It seems the only difference is its deeper. The "I don't know when to say when" quote sounded like strong mad inhaled helium. The real strong mad's voice is much deeper, and I believe strongbad's is usually deeper as well.

Something's definitly much different.

Remember Decemberween 2004, how the voices sounded off in that? I think that was the same deal as the quotes of the week. I had some technical answer, but I forgot >.>

_________________

Oh this is a kind of special case: it is an outtake from the 'isp' sbemail, so we don't have an original to compare with; and if you remember strongmad talks with a "high" pitched voice in the final so the normal voice of the character is purposely modified.

I forgot to mention the effect of the lossy mp3 encoding at a very low bitrate (16 kbps in sbemails): this could heavily modify the sound too.

Anyways: if you resample the quotes to 11 kHz and then encode to mp3 at 16 kbps, you'll get the "usual" voices. No speed changes involved.

I see you live up to your namesake, Nerdbert.


Seriously, thanks. I've never actually noticed this phenomenon myself, but you may be right about all this.

_________________
So, so you think you can tell Heaven from Hell, blue skies from pain. Can you tell a green field from a cold steel rail? A smile from a veil? Do you think you can tell?

Ohh, I totally thought something like that was going on! Thanks for the highly technical explanation; I've been waiting for it for a while.

By the way, you can distinctly hear a brief moment in the 2004 Decemberween toon when Homestar's voice is downsampled: It's when he trips and falls in front of Marzipan ("Oof!"). I think this is because it is used in conjunction with a sound effect... do you think that is the reason, Nerdbert? Or it could have been overlooked.

_________________
dot dot dot

Um... I really can't notice this, I mean: the whole audio track is encoded consistently with the other toons' ones: plain ol' mp3, 16kbps @ 11kHz. That is: it's wholly downsampled.

BTW, don't take what I said (even earlier) for the truth, it is just what I think it most likely happened.

I can see what you mean, although it doesn't sound THAT different to me. Perhaps they realised they didn't have an "oof", but couldn't be bothered to record a new one. It is a pretty generic noise after all.

By the way, am I right in thinking that lower-frequency sounds are less noticeably distorted when compressed?