London When pop and rock bands practice or perform, they rely on their guitars, keyboards and drumsticks to make music. Oliver McMann, a British AI music maker who goes by the stage name Imoliver, dropped his chat.
McCann’s songs span a range of genres, from indie-pop to electropop to country rap. There’s only one major difference between McCann and traditional musicians.
“I have no musical talent at all,” he said. “I can’t sing, I can’t play instruments, and I have no musical background at all.”
McMann, 37, who has a background as a visual designer, began experimenting with AI to see if it could boost his creativity and “bring some of my words to life.” Last month, he signed with independent record label Hallwood Media after one of his tracks reached 3 million streams, which is billed as the first time a music label has signed an AI music creator.
McCann is an example of how Chatgpt-style AI song-generation tools like Suno and Udio have spawned a wave of synthetic music, a movement most notably highlighted by the fictitious band Velvet Sundown, which went viral despite AI creating all of its songs, lyrics, and album art.
It fueled debate about the role of AI in music, while raising fears about “AI Slop” – the automatic generation of low-quality mass content. It also brought attention to AI song generators that are democratizing songmaking but threatening to disrupt the music industry.
Experts say generative AI is set to transform the world of music. But so far, there are few details about how it will impact the $29.6 billion global recorded music market, which includes about $20 billion from streaming.
The most reliable figures come from music streaming service Deezer, which estimates that 18% of songs uploaded to its platform each day are AI-generated, though that only accounts for a small portion of the total number of streams, suggesting that few people are actually listening. Other, larger streaming platforms like Spotify haven’t released any figures on AI-generated music.
Udio declined to comment on how many users it has or how many songs it has created. Suno did not respond to a request for comment. Both have free basic tiers, as well as Pro and Premium tiers that have access to more advanced AI models.
“It’s a complete boom. It’s a tsunami,” said Josh Antonuccio, director of the School of Media Arts and Studies at Ohio University. The amount of music created by AI is “only going to increase exponentially” as young people grow up with AI and become more comfortable with it, he said.
However, generative AI, with its ability to spit out seemingly unique content, has divided the music world, with musicians and industry groups complaining that recorded works are being used to train AI models that power song generation tools.
Record labels are trying to address the threat that AI music startups are creating their own revenue streams even as they hope to use it for new income, while recording artists worry that it will devalue their creativity.
Trīs galvenās ierakstu kompānijas, Sony Music Entertainment, Universal Music Group un Warner Records, Iesniegti tiesas prāvas Pagājušajā gadā pret Suno un Udio par autortiesību pārkāpumiem. Tiek ziņots, ka jūnijā abas puses arī uzsāka sarunas, kas varētu pārsniegt tiesas prāvu nokārtošanu un noteikt noteikumus par to, kā māksliniekiem tiek maksāts, kad AI tiek izmantots viņu dziesmu remiksēšanai.
Vācijas honorāru kolekcijas biedrība Gema ir iesūdzējusi tiesā Suno, apsūdzot to par Lou Bega un Alphaville dziesmu, piemēram, Mambo Nr. 5 ”ģenerēšanu un Alphaville“ Forever Young ”.
Vairāk nekā 1000 mūziķu, ieskaitot Keitu Bušu, Anniju Lennoksu un Damonu Albarnu, izlaida a Klusais albums Lai protestētu pret ierosinātajām izmaiņām Lielbritānijas likumos par AI, viņi baidās, ka viņu radošā kontrole iznīcinās. Tikmēr šo tehnoloģiju ir izmantojuši citi mākslinieki, piemēram, Will.i.am, Timbaland un Imogen Heap.
Daži lietotāji saka, ka debates ir tikai vecu argumentu atkārtošana par vienreiz jaunu tehnoloģiju, kas galu galā tika plaši izmantota, piemēram, autotūna, bungu mašīnas un sintezatori.
Cilvēki sūdzas, ka “jūs izmantojat datoru, lai veiktu visu darbu jūsu vietā. Es to neredzu tā. Es to redzu kā jebkuru citu rīku, kas mums ir,” sacīja Skots Smits, kura AI grupa Pulse Empire iedvesmoja 1980. gadu britu sintezatoru grupas, piemēram, New Order un Depeche režīms.
Smits, 56 gadi un daļēji pensionēts bijušais ASV Jūras spēku sabiedrisko lietu virsnieks Portlendā, Oregonā, sacīja, ka “mūzikas producentiem ir daudz instrumentu viņu arsenālā”, lai uzlabotu ierakstus, kurus klausītāji nezina.
Tāpat kā Makmens, Smits nekad nav apguvis mūzikas instrumentu. Abi saka, ka viņi daudz laika un pūļu pieliek savu mūziku.
Tiklīdz Smits gūst iedvesmu, dziesmu tekstu uzrakstīšana viņam prasa tikai 10 minūtes. Bet tad viņš pavadīs pat astoņas līdz deviņas stundas, veidojot dažādas versijas, līdz dziesma “Satriecina manu redzējumu”.
Makkanans sacīja, ka viņš bieži izveidos līdz 100 dažādām dziesmas versijām, pamudinot un atkārtoti iespaidot AI sistēmu, pirms viņš ir apmierināts.
AI dziesmu ģeneratori var izspiest dziesmu tekstus, kā arī mūziku, bet daudzi pieredzējuši lietotāji dod priekšroku rakstīt savus vārdus.
“AI dziesmu teksti mēdz iznākt diezgan klišejiski un diezgan garlaicīgi,” sacīja Makmens.
Lukas Rams, Filadelfijas apgabala iedzīvotājs, kurš veido dziesmas savai AI grupai, kas guļ ar vilkiem, sacīja, ka AI dziesmu teksti parasti ir “īpaši kukurūza”, nevis tik radoši kā cilvēks, bet var palīdzēt sākt rakstīšanas procesu.
“Tas darīs ļoti pamata atskaņu shēmas, un tas turpinās atkārtot to pašu struktūru,” sacīja Rams, kurš raksta savus vārdus, dažreiz, ieliekot savus bērnus gulēt un gaidot, kad viņi aizmigīs. “Un tad jūs iegūsit vārdus tur, kas ļoti stāsta par AI ģenerētiem dziesmu tekstiem, piemēram,” Neon “, jebko ar” ēnām “.”
Rams used to play drums in high school bands and collaborate with his brother on his songs, but work and family life began to take up more of his time.
Then he discovered AI, which he used to create three albums for Sleeping with Wolves. He’s taking it seriously, creating CD Jewel cases with album art. He plans to put his songs, which blend Metalcore and EDM, more widely online.
“I want to start putting it on YouTube, public or for distribution or whatever, just to have it out there,” Ram said. “I might as well, otherwise I’m literally the only person who hears this thing.”
Experts say AI’s potential to allow anyone to come up with a hit song is poised to shake up the music industry’s production pipeline.
“Just think about what it used to cost to make a hit or do something that breaks,” Antonuccio said. “And it just keeps shrinking from a big studio to a laptop to a bedroom. And now it’s like a text prompt – multiple text prompts.”
But he added that AI music is still in the “wild west” phase due to a lack of legal clarity around copyright. He compared it to the legal battles more than two decades ago with file-sharing sites like Napster, which heralded the shift from CDs to digital media and ultimately paved the way for today’s music streaming services.
The creators also hope that AI will eventually become part of the mainstream music world.
“I think we’re entering a world where anyone, anywhere, could make the next big hit,” McMann said. “As AI becomes more widely accepted among people as a musical art form, I think it opens up the opportunity for AI music to appear on the charts.”
