The VOCES8 Foundation presents its latest Live From London digital festival this weekend which, among other things, will include the premiere performance of a new piece composed by Ed Newton-Rex that features lyrics generated by artificial intelligence.
Composer and entrepreneur Newton-Rex has been involved in generative AI in the music space for more than a decade, having founded the music AI start-up Jukedeck that was subsequently acquired by TikTok owner Bytedance. He is now VP of Audio at Stability AI.
For his new composition – called ‘I Stand In The Library’ – he used OpenAI’s GPT-3 to generate a set of lyrics which he then set to music. It will be performed by the VOCES8 Choir in a Live From London concert that will be available to stream on-demand from Saturday.
Newton-Rex spoke to CMU’s sister media ThisWeek Culture about the new composition, explaining: “When OpenAI’s GPT-3 was released to initial testers, I heard about it and signed up for early access. In the early days lots of people were sharing poetry it had written them, some of which was surprisingly good – so it wasn’t a great leap to think perhaps I could use it to write a text for my [next music project]”.
He initially prompted the generative AI with the line “below is a poem about music and solitude”. He goes on: “When I hit ‘enter’, it immediately came out with the line ‘I stand in the library, where a voice soars’, which I loved”.
With a little more work, the AI generated a complete poem which Newton-Rex then set to music. “The generated text directly influenced the music in a major way”, he says. “The mention of a piano in the first stanza led me to write for choir and piano, something I haven’t done much before – I tend to write for unaccompanied choir”.
While Newton-Rex has been working in the generative AI domain for many years now, it is obviously in the last year that the potential of music-making AI has become a really big talking point within the music industry. Though, he points out, while AI generated the words for his new piece, it couldn’t have created the musical element. Well, not yet.
“Most state of the art AI music generation systems today create short passages of audio, perhaps fifteen to 60 seconds, and tend to do much better at instrumental music than vocal music”, he says. “There is nothing yet that comes close to creating a structured fifteen minute piece, setting text effectively, creating something musically novel in the process – which is hopefully something I’ve managed with this piece, though that’s for the listener to decide”.
“But will it be able to in future?”, he goes on. “Absolutely – or at least it’s highly likely. In other modalities, like image and text, the speed with which generative AI systems have recently improved and achieved things that two years ago seemed impossible is astounding. We should expect the same to happen in music”.
Read the full interview – including Newton-Rex’s thoughts on the relationship between music-makers and music-making AI – on ThisWeek Culture here.
There is more information about the Live From London concert in which the premiere performance of ‘I Stand In The Library’ will appear here.