ALAC comes from Apple Lossless Audio Codec, is a reference decoder for Apple's lossless audio codec. The decoder has been written by reverse engineering Apple's proprietory audio format.
Located here is a basic decoder for Apple Lossless Audio Codec files (ALAC). ALAC is a proprietary lossless audio compression scheme. Apple never released any documents on the format.
What I provide here is a C implementation of a decoder, written from reverse engineering the file format. It turns out that most of the algorithms in the codec are fairly well known. ALAC uses an adaptive FIR prediction algorithm and stores the error values using a modified rice or golumb algorithm. Further details are in alac.c.
Although an encoder is not provided, by using the decoder as a sort of specification it should be fairly trivial to write an encoder. By exploiting other lossless audio encoders, such as FLAC, the task will be much easier. Although one wouldn't be able to copy the compression algorithms verbatim, as adaptive compression is used in ALAC and not in FLAC. There are, however, a bunch of academic papers on the issue.
The program located here will not be able to handle all ALAC files, it can only handle mono or stereo files. ALAC allows up to 8 channels. It should be trivial to finish the implementation once I find files that I can test it with. Likewise the decoder only supports 16bit sample sizes. Again, it should be trivial to fix.
The decoder is fairly self explanatory, it can read an ALAC stream from either a file or from stdin, and write it as raw PCM data or as a WAV file to either stdout or a file. In theory one should be able to stream data to the decoder.
What's New in This Release:
Comments not found