32 lines
1.4 KiB
Mathematica
32 lines
1.4 KiB
Mathematica
|
## Author: Ramesh Yerraballi, Jonathan Valvano
|
||
|
|
## Use this in octave to convert a wav file into raw binary
|
||
|
|
## samples with a chosen precision at 11.025kHz
|
||
|
|
## You will need to load the signal package first to use the decimate function
|
||
|
|
## You can load it like this on the octave prompt:
|
||
|
|
## pkg load signal
|
||
|
|
## Run in octave as:
|
||
|
|
## WCbin ("filename", precision);
|
||
|
|
## The inputs are (i) the name of the wav file without the extention .wav.
|
||
|
|
## (ii) the sample precision in bits (e.g., 4 to 8 bits)
|
||
|
|
## This will generate a filename.bin file that has raw binary data
|
||
|
|
## For example, say you have a sound file called boom.wav that you want to use in
|
||
|
|
## in your game.
|
||
|
|
## 1. Put the WCbin.m script and the boom.wav file in a folder (say sounds).
|
||
|
|
## 2. Open octave/matlab and cd to the sounds folder
|
||
|
|
## 3. Run script as: WCbin("boom", 8)
|
||
|
|
## For 8-bit precision
|
||
|
|
## This will produce a file named boom.bin, which contains raw samples
|
||
|
|
## You can put the file on an SDC and read the data at run time
|
||
|
|
function WCbin(filename,precision)
|
||
|
|
# load the file
|
||
|
|
[SplsOrig, fs] = audioread(strcat(filename,'.wav'));
|
||
|
|
# downsample to 11.025kHz
|
||
|
|
Spls = decimate(SplsOrig,round(fs/11025));
|
||
|
|
# trim the precision of each sample
|
||
|
|
Spls = round((Spls+ 1)* (2^precision-1)/2);
|
||
|
|
# write raw binary to file with bin extension
|
||
|
|
file = fopen(strcat(filename,'.bin'), 'w');
|
||
|
|
fwrite(file, Spls(1:length(Spls)));
|
||
|
|
fclose(file);
|
||
|
|
end
|