An audio language is used to generate audio sources for an audio scene. The audio sources are patterns comprising other patterns, composed music, audio samples, instruments, MIDI instruments, and filters. The audio files can be located on a computer-readable medium or accessible over a network. An audio player is provided for executing audio sources contained in audio files having code segments for the audio sources written in the audio language. The audio player prioritizes the audio sources according to prioritization reserved variables of the audio language to determine an order of priority. The audio player executes the audio sources according to the order of priority by transferring, playing, storing, or any combination of transferring, playing, and storing the audio sources according to the order of priority of the audio sources. The audio language and the audio player are advantageously useful for executing audio sources for an audio scene in a virtual environment.