Microsoft recently filed a patent on the WIPO Intellectual Property Portal entitled “Artificial Intelligence Models for Synthesizing Audio.
The patent states that Microsoft is inventing an intelligent audio synthesis technology for creating sounds, music and other audio elements for various media such as movies, TV shows, games, and even live recordings. The patent refers to dynamic moments in games, showing that it can create scores that match the player’s actions. The abstract of the patent states that visual, audio and textual features and cues (collectively referred to as “datasets”) can be used to set parameters that guide numerous artificial intelligence models in building audio scores.
Artificial intelligence plays an integral role in video games. From enemy behavior and combat encounters to procedural level generation and interaction with NPCs and environments, AI is integral to all levels of game development. In terms of sound design, adaptive/dynamic soundtracks are featured in many video games such as the modern Doom games, Alloy Gear Rising, Ghostbusters 5, and more.
Microsoft’s new patent description details numerous artificial intelligence engines that are tasked with audio playback based on the data sets provided. They can analyze human expressions and emotions, collect location data, analyze the tone of a situation, and more. The AI can learn pictures, videos, movies, and live events and produce a set of audio files that can be layered with appropriate sound effects and music for visuals. This high-tech artificial intelligence could open up many exciting paths for media creation. One could use a huge and growing library of audio scores for movies, games, etc. For example, designing an epic orchestral piece for a hero’s appearance, composing a melancholy tune for a pet’s departure, and developing sound effects for gunshots and explosions, all of which could be entrusted to the AI’s algorithms.
The technology will be driven by cloud computing. When the system will actually be available remains to be seen, but the future of audio design looks promising, and Microsoft could be at the helm of a revolution in this area.
