1.       The soundtrack, the music, has for main purpose to transmit sensations to the player, to make the player feel sad, feel energy, feel like laughing, to feel “pumped up” ,to feel surprised . . .

2.       The sound effects are done to inform the player, to make the player know something by audibly saying it, but not necessarily in the most obvious way. It is to make the player hear a sound and think of coins instead of having a voice saying “you just got +30 coins”, “you just got +120 coins”, “you just got a lot of coins”.

3.       Background audio is almost the fusion of sound effects and music. They intend to make you feel like you are inside the game. At the same time, they attempt to inform the player “you are on a Caribbean beach” and make the player feel like the player is on a Caribbean beach. They both inform the player "you are IN SPACE" and create a sensation of immersion by making the player FEEL like he is, like the example, IN SPACE! 

They make the player merge with the game’s world, to make it more immersive, to make it feel like the player is a part of the game and/or the game’s world.

Besides their purpose, there are other differences between them, with 2 main differences: length and “trigger”.

1.       Sound effects usually last a few seconds, less than 7 seconds and are TRIGGERED by something (steps, shooting, level up, but it won’t simply “pop up”).

2.       Background sounds usually last longer, but not so much longer, usually lasting more than 4 seconds but less than 15 seconds. They are “randomly triggered”, they are usually there, but they usually do not “overlap” one another, and they might “not be there”, despite their “triggers” being there.

3.       Music usually last a few MINUTES and is usually “constantly there”, with no moments where there is no background music at all.  

Procedural audio is a method of creating sounds in real-time, often using forms of synthesis such as physical modelling, whilst generative music creates an evolving soundscapes, controlled by the game according to a set of rules. These techniques enable a wider range of variability and deeper game integration than would be possible using more traditional linear audio and can lead to a more varied, unpredictable and believable soundscape.

Paul Weir is the Audio Director for Hello Games No Man’s Sky, an innovative space action, adventure and exploration game. In this talk, Paul will look at how the game sound designer is being challenged by systemic gameplay techniques and the ways in which the technology, and more importantly, the creative skills, are evolving. He will discuss No Man’s Sky procedural creature vocalisation system and their generative music tool and the challenges involved in building these new applications.

http://sonarplusd.com/activity/never-ending-audio-generate-sound-games-paul-weir/

According to http://www.vice.com/en_uk/read/no-mans-sky-sounds-amazing-literally-514 No Man's Sky's music was partially randomly generated, as the procedural sound system picks certain "bits" of sound and "match" them depending on the circumstances of the gameplay.

Perhaps the most impressive use of audio in No Man's Sky is its animal's procedural sound system, which according to http://www.newyorker.com/tech/elements/what-a-dragons-mating-scream-sounds-like is created by simulating a pharynx, vocal cords and a mouth for each animal to make them sound different.