I am working on a project which I
sort of plan to add voice acting. I don't have any real voices, so I'm going to be using text-to-speech. (The characters who are speaking are supposed to be robotic anyways.)
Although, I'm transferring to the 3DS just by playing it and having the 3DS record it via a TRS cable, and playing it back using WAVSETA.
You can easily record 30 seconds of audio just by running MICSTART 0,0,30. In fact, you can record up to record 128 seconds with this. It limits you to 32 seconds, but if you throw the waveform data into an array, then resize the array every 32 seconds and add the new waveform data to it, you can get to like 128 seconds before you run out of memory.
I do have the audio DLC as 12Me21 mentioned, but most people don't have it, and it's only two short subroutines, ~20 lines to play audio files in the background with WAVSETA, so there's no point in restricting users just for slightly higher audio quality when it's so easy to implement audio playback without it.
In fact, here are the 19 lines I've been using to play audio files back.
Spoiler
@LOADWAV
DIM AUDIO[0]
FINISHED=FALSE
CLOCK=MILLISEC
PROGRESS=0
LOAD "DAT:"+FILENAME$,AUDIO,FALSE
RETURN
@PLAYBACK
FIN=FALSE
IF MILLISEC>=CLOCK AND LEN(AUDIO) > 8180*PROGRESS THEN
WAVSETA 250,127,0,127,127,AUDIO,78,8180*PROGRESS,8180*PROGRESS+8179
BEEP 250
PROGRESS=PROGRESS+1
CLOCK=MILLISEC+1000
ELSEIF LEN(AUDIO) <= 8180*PROGRESS THEN
FINISHED=TRUE
ENDIF
RETURN
You can use that to play audio files by first setting "FNAME$" to the WAV file's name, then GOSUB'ing @LOADWAV, then adding "GOSUB @PLAYBACK" to the inside of your game loop. (It only plays audio files recorded in 8-bit fixed at 8180 Hz.)
Given, the quality is pretty low. You can use EXTFEATURE to determine if the person has the DLC, so it might be a good idea to have the high and low quality audio and choose which to play based on whether the user has DLC.
