Mystery Dungeon sound effects?

you could download and chop up these:

Rescue Team sound effects
Explorers sound effects
Gates to Infinity sound effects

The sound sample aren't encrypted, they're just encoded as 4 bits ADPCMs. It could be using the standard nintendo DS ADPCM encoding, or a proprietary format.

I haven't found the proper decoding table they're using to decode the sound yet. But by running the game in the no$gba debugger and isolating where sounds are loaded, and looking at the assembler code, someone should be able to figure out those. Its just pretty time consuming, and its really all about trial and error.

Then, most sound effects for moves and such are basically sequences made by playing the sound samples in a certain order with varying pitches and etc. Those sequences are stored in the /SOUND/SE/ folder in Explorers of Sky, as SED files(SEDL). Moves sound effects are named "waza" which is the romaji form for the Japanese word for art/technique : jisho.org/words?jap=waza&eng=&dict=edict
SED are very similar to the SMDs used for the game's music, but there are most likely differences.
Though, on my side of things, with the research, music/sound is relatively low priority for now, given that there weren't any big breakthrough made on that side of the game for a while.. Figuring out the ADPCM tables would seriously help a ton though. However, I do know that, you can get relatively good decoding of the samples by using the raw ADPCM decoder in audacity.
However, you gotta cut out the sound data with an hex editor and put it in its own file for that to work properly, because ADPCM basically bases each 16 bits samples it decodes on the value of the lasts, so feeding it random bytes can't be really good..

Also, here's the toolset the devs used for the music and sound effects in the Explorers of Sky/Time/Darkness games : www.procyon-studio.co.jp/dse/spec.html
If someone was able to translate what's in some of those pictures, it could help give a better idea of what might be stored in those SED files.

No problems
Its probably the only thing I can help with around these forums anyways XD

But don't think PCM and ADPCM are the same thing.
PCM is a bunch of raw samples most of the time stored as 16 bits, and ADPCM is a bunch of sample encoded on 4 bits instead of 16.

Samples are fairly easy to understand though.
Let's take a look at mono, 44,000hz , 16 bits PCM data for example. That's what's used in most WAV files really, its as basic as it gets.
Its literally 16 bits integers one after the other. The sample rate, 44,000hz, literally only means that there are 44,000 such sample for every seconds of that sample playing.
As for what the data represents, some article I read in the past had a nice way to put it. You can think of it as if each of those integers is the strength of the electric current sent to your headphones when listening to that audio data. The membrane that makes the sound would change position 44,000 times a second(in a perfect world), and in turn reproduce the waveform that was recorded into the file's PCM data.

ADPCM on the other hand, is basically a way to compress that same PCM data into only 4 bits ! That's literally only enough to count from 0 to 15.. And it does that at a pretty big cost in audio quality! I'll admit I'm not 100% sure what's the general idea behind it, even after reading the official specs for the format.. But decoding it involves applying a mathematical formula on each 4 bits sample, where the value is compared against a table. And then at the end of that process, you get a 16 bits integer / 16 bits PCM sample.

And thanks for offering help with those ! I'd have translated them myself, but it takes you much longer when you're not fluent at all, or can't even write/read Japanese XD

Holy nuts, this is absolutely perfect; I can't thank you enough for this. It works like a charm!
_{my 8 month quest has been completed}