How to change ProffieOS built-in sounds (e.g. "SD card not found")

I wanted to change some of the default sounds in ProffieOS. Most notably ones such as “SD card not found,” “Font directory not found,” “Font directory too long” and create messages that would make more sense in-galaxy. However, I couldn’t find any wav files, or any kind of files which might be parsed in compiling as sounds. Where does ProffieOS get these sounds? Is it possible to change them?

They are built-in “talkie” sounds using the voice_data, not wav files. If your SD card is not found or there’s an issue with the font then there’d be no wav file to play.

Ya, that makes sense. So then how would I go about editing those messages? These are hexadecimal values right? What do they reference?

I use BlueWizard on MacOS to convert 8bit wavs to hex.
The Windows version is called QBox Pro. Not sure if you can get it to run on modern Windows, it was built for Windows 3.1 and prob won’t work past XP.

You’d edit the file Fett263 posted above with the new hex string, leaving the error code, so that even though the OS is calling for “talkie_font_directory”, it will play the hex you pasted instead.



You’ll need to read the posted links and follow external links to the executables.

It’s not an easy road to take, but it does work.

Since I already have it setup to work, you can toss the wavs to me and I can generate the hex for you if you want :wink:

Well, I’ve got a Hackintosh, so it should work. wdym by “leaving the error code” Thanks for providing these links btw. Time to jump into yet another rabbit hole!

Also, thanks for offering! I would like to try and figure it out, but if not, I might send it over to you, lol.

I guess I should also ask what values exactly I should replace. I assumed the ones at the very bottom “talkie_sd_card_15” etc.

Oh if you’re Hackintoshing you’re a perfect candidate. lol
“Leaving the error code” meaning you need to leave the part that’s called for:
const unsigned char talkie_font_directory_15[] = {
and just edit the hex string.
Tips: leave a bit of silence at the front end of the wav.
I can’t think of other tips at the moment, but adjusting a slider or 2 might help it sound better.
Here…take this instead of hunting for the app:

Ah, ok. I think that makes sense, we’ll see when everything goes wrong :stuck_out_tongue:

I already got BlueWizard compiled and running actually, thanks though!

Thanks for the tips!

Yeah, it’s going to piece those together to make sentences to speak.

1 Like

Another quick question as I am messing with this, You said you high and low pass the audio, what values do you use for that?

I didn’t say I do that for this process. I would suggest maybe high pass at ~100hz, but keep the high end. A ton of it will go away on it’s own as you export it as only 8kHz.

1 Like

Tip. You can test what it sounds like from the board without having to upload by using serial monitor. Once the hex is generated in BlueWizard,
Type talkie and paste the hex then send.
It might be talkie_15 to play it at the speed the board actually will.
That’s a speed value. Notice it in all the lines in voice_data.h.
Playing further:
You can play with the speed values.
You can replace anything the board says such as numbers etc…, although it might make things weird.
Search ProffieOS folder for any occurance of talkie to see what gets called for.

Huh, that’s really interesting! Thanks! One question I do have is that after running it through BlueWizard, at least in it’s preview, the audio has a lot more noise behind it, and the tone is completely off… any ideas what might be causing that? Or is that just the nature of this?

click on the 2 tick boxes:
“include hex prefix (0x)” and “include explicit stop frame”

Ya, I have.

It’s not pretty, and not easy. Source can’t have any music or background stuff. Pure vocals or else you’ll prob get a mess.
Keep playing around, but send me the files too to try and see if they’ll work

Ok, will keep moving levers!

This is the clip I am currently working on after I have ran it through Audacity to get it to 16 bit 8000 Hz.

haha. Zarvox

You recognized it! That’s impressive, lol!

lol. It’s currently set as my current voice no joke.
I’m a Mac dude since OS 8.5 .