Although “It Could Be Worse” and “Tales of the Atlantic,” two songs spontaneously generated by a powerful algorithm in under a minute, exhibit polished production, prominent drum presence, and appealing lyrics.
Bob Hallett, a founding member of the folk-rock ensemble Great Big Sea from Newfoundland, expressed his reservations about the AI-created tracks. He pointed out that the genre attribution as a country singer was inaccurate, and the lyrics lacked proper rhyming schemes, resulting in an overall peculiar sound.
Upon listening to the generated music, which utilized Suno’s AI technology trained on Celtic, Folk, Loud, and Passionate Great Big Sea elements, Hallett rated the composition a mere two out of ten, deeming it a far cry from the band’s signature style.
However, experts foresee an upward trajectory in the quality of such AI-generated music. Jimmy Lin, an educator and chairperson at the University of Waterloo specializing in artificial intelligence, highlighted the rapid advancements in technologies like Suno, indicating a continuous improvement in their output.
Suno is among the companies at the forefront of developing AI software for creating original songs based on textual prompts. Users can craft songs either from instrumental tracks provided by the program or customer-generated inputs. However, the system refrains from composing music using copyrighted phrases, ensuring originality and avoiding legal complications.
Contrary to controversial instances like “Heart on My Sleeve,” an AI-generated composition blending the vocals of Drake and the Weeknd, Suno’s creations possess a reflective Celtic essence when tasked with producing a “Great Big Sea song about cod fishing.” Yet, they fall short of replicating the authentic sound of the band.
Another comparable program, MusicFX by Google, bears similarities to its AI Test Kitchen platform. Additionally, Adobe recently unveiled Project Music GenAI Control, an early-stage AI audio creation and editing tool.
In December, Microsoft integrated a Suno-powered song generator into its Copilot robot, enhancing its conversational capabilities with users through music.
Lin explained that these AI programs are trained using algorithms to predict subsequent elements based on extensive data sets. Music-generating algorithms predict acoustic sequences, while chatbots anticipate the next word in textual responses.
Addressing concerns about potential copyright infringement, Lin emphasized the importance of obtaining consent and compensating musicians whose work is utilized to train AI platforms.
Despite the uncertainties surrounding AI-generated music, Hallett remains unfazed, acknowledging the challenges faced by artists in an era dominated by digital platforms like YouTube and Spotify that have reshaped the music industry landscape.
While AI-generated songs may cater to commercial needs, Hallett believes that genuine human connection through music remains unparalleled in resonating with audiences on an emotional level.
In conclusion, the debate over AI’s role in music creation continues, with the balance between technological innovation and artistic authenticity at the forefront of discussions within the industry.
