The music industry has been abuzz with speculation about what, exactly, Facebook has planned for music beyond its first set of user-generated-content focused licensing deals.
Few people have suggested that the social network’s agenda includes using artificial intelligence for “translating music across musical instruments, genres, and styles”.
However, that’s exactly what a team within Facebook’s AI Research department has been doing, according to a paper setting out their work on a “universal music translation network”, which was spotted by tech-news site TheNextWeb.
What’s that? “In this work we are able, for the first time as far as we know, to produce high fidelity musical translation between instruments, styles, and genres,” explain the researchers. “For example, we convert the audio of a Mozart symphony performed by an orchestra to an audio in the style of a pianist playing Beethoven…”
“Our results present abilities that are, as far as we know, unheard of. Asked to convert one musical instrument to another, our network is on par or slightly worse than professional musicians. Many times, people find it hard to tell which is the original audio file and which is the output of the conversion that mimics a completely different instrument.”
We’ll let you dig in to the technical details via the paper, if you so desire. But there’s an interesting line on what this research could lead to: “Our work could open the way to other high-level tasks, such as transcription of music and automatic composition of music.”
Music Ally has been covering the intersection between AI and music regularly. Read our report on the AI day at the recent The Great Escape conference, for example, or our coverage of a panel on the topic at AIM’s Music Connected event. Or, indeed, the panel we ran at the by:Larm conference earlier this year.