Artificial Intelligence in Music and Audio

Image: –

Artificial Intelligence (AI) is the field of computing which creates machines that can operate in human-like ways, with the ability to ‘think’, to learn and improve their own capabilities based on often vast amounts of data. AI is being brought into many aspects of our lives, the world of music and audio is one of them.

There have been fascinating developments in AI applied to the creation, recording, distribution and presentation of music to audiences. The software can be viewed as a creative collaborator with composers and musicians, as a set of tools to enhance their creative visions and help them realise them, and as a creator of original music itself. AI software can function as a mastering engineer to ‘polish’ recordings for release. Streaming services use AI to build profiles of their users and suggest songs, artists and playlists that their audience will enjoy.

There are web-based services such as Amper, which require simple input of a few parameters – genre, tempo, duration – to create original music for a user’s requirements. The music can be tweaked in a variety of ways including changes in instrumentation and feel, until it meets the user’s need. This is useful for YouTube content creators needing music for their channels. The AI is capable of more: the singer Taryn Southern has made an album which she wrote using Amper AI. It has been estimated that in the next decade, between 20% and 40% of all top 40 hit singles will be written partially or entirely by AI software.

Video: Taryn Southern

When music has been recorded, it needs to be mastered to achieve the best sound for replay media. This is a highly skilled time-consuming job requiring experienced and expensive specialist mastering engineers to perform. AI services have been introduced to master recordings quickly and cheaply. The best-known online mastering service is Landr ; Abbey Road Studios owns its competitor, Cloudbounce . Recordings are uploaded to the services, which use AI algorithms to analyse the musical and technical qualities of the recordings, then process them using eq, dynamics processing, level adjustments and other mastering techniques. The resultant mastered recordings are perfectly adequate for most listeners. Although the artistic touch of a human mastering engineer is not present, the AI services are continuously learning and improving.

Radio stations use programme-scheduling software that constructs playlists matched to their audience demographics. Streaming services such as Spotify and Apple Music implement AI to construct detailed databases of their users’ tastes in music. The data is used in sophisticated processing to curate playlists and to continue to play music after a selected album or playlist has finished, so users continue to be entertained and are simultaneously introduced to new artists and music they may enjoy. The accuracy of the music choices and the audience’s positive reception of them are testament to the success of the use of AI in this instance.

The music industry is a complex combination of art and humanity, technology and skill, marketing and entertainment. AI is becoming intrinsic to the way we will write, record, distribute and enjoy music, now and in the future.

Sennaar K (2019) 6 Applications of AI for Audio

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: