Eclipse Nkasi created an entire music album in three days with AI and thinks everyone else should too

To show that AI might create compelling music on a funds, Eclipse Nkasi, created an AI music album in three days with lower than $500. 

A number of months in the past, AI-generated photos might simply be recognized by the awkward drawing of human arms. At the time, many mocked AI’s lack of ability to attract arms and questioned the declare that it might exchange creatives. But AI-generated photos have improved, and these days, it may be troublesome to distinguish between photos produced by AI and these created by human artists. The identical is occurring with music, as AI can now mimic the voices of widespread artistes and sing in their types.  Despite these developments, many nonetheless doubt that AI can produce good music. 

Independent artist and former promoter for Nigerian document label Chocolate City Music, Nkasi, tells TechCabal, “When it came to music, people were very sensitive or emotional about their stance with AI. People felt it wasn’t possible that AI could originally create art. You hear arguments like AI lacks soul or it is not possible to generate anything close enough. I took that as a challenge.” 

While AI music turbines can produce spectacular and distinctive sounds throughout numerous genres with generic or detailed prompts, there aren’t as many able to producing spectacular authentic songs. Models like Google’s MusicLM, which was opened to the general public yesterday, could make attention-grabbing instrumentals. But these fashions stumble on the subject of lyrics. The lyrics of music produced on MusicLM months in the past have been reportedly jumbled and incoherent, just like how phrases will be scrambled on photos generated by OpenAI’s picture generator Dall-E. This model that was publicly launched won’t generate songs with the voice of somebody singing, as which will appeal to copyright issues similar to a lot of the songs that includes the voices and music of identified artists have not too long ago. Spotify not too long ago took down 1000’s of AI-generated songs resulting from copyright issues.

Read additionally: What will occur now that AI can sing like your favorite Nigerian musician?

Something authentic and extra reasonably priced

Most of the favored AI-generated songs function the AI-cloned voices of widespread artists singing already present songs, not new ones. Remember the AI-generated music that includes Drake that went viral in April? That was not an authentic. It was a rap acapella reworked with an AI-generated Drake voice. The new vocals have been positioned over an already-existing beat that had been modified. Nkasi needed most if not each, inventive component of the album to be initially created by AI, so he used completely different AI instruments to work on every half—lyrics, instrumentals, and singing—and put them collectively. 

Not solely did Nkasi wish to show that AI might create authentic and emotionally compelling music, he additionally needed to indicate that AI might do it on a good funds. This ambition was motivated by the previous Chocolate City music promoter’s curiosity and his expertise dealing with the music enterprise as a part of a label and as a solo artist. “Music is very expensive to create and promote, especially for new artists,” Nkasi informed TechCabal.  According to him, a music album venture typically prices ₦2 million ($4,339) to ₦3 million ($6,509) for an common artist. “You need to pay for beats and the average music producer in Lagos who knows what they are doing would charge you about a ₦100,000. Producers like Masterkraft charge well over a million naira. If you want to do this on a basic level, ₦100,000–₦200,000. If you are going to do 13 songs, you can do the math. Also, mixing and mastering can cost around ₦100,000.”

But Nkasi needed to create an album for lower than $500 (₦231,000) in three days. The results of his efforts is an album titled Infinite Echoes, which he printed on streaming platforms like Spotify and YoutubeMusic underneath his studio title, Eclipse Nkasi. “Apart from the overheads of generators and so on, it was insanely cheap. I don’t think I spent more than half of the $500 budget I had. The most money I spent on any one tool was the voice synthesis tool that allowed me to create an AI singer that  I christened Mya Blue. That cost me about $168 for the software and voice bank. Extra voice banks cost around $70,” Nkasi mentioned.

Read additionally: Google is about to alter the way you create and take heed to music with its AI music generator

Creating the album

Save for some enter from him, Nkasi insists that the entire inventive course of of manufacturing the nine-track album was finished with synthetic intelligence. According to him, AI steered the title, “Infinite Echoes.” It additionally conceptualized the album, created the album cowl, wrote the songs, and created the tracklist, music variations, and instrumentals. AI additionally sang a lot of the songs. “I am well aware of what it takes to create an album on the artist and business sides, but I just wanted to get the album done in three days,” he informed TechCabal in an interview.

Nkasi developed the album’s instrumentals on an AI music generator, SoundRaw. The generator has a financial institution of royalty-free instrumentals generated by AI. “You can tailor the length, tempo, composition, instruments, and genre of the sounds to suit your needs,” he mentioned, including that he has a paid account that prices $15 a month. He downloaded a lot of the instrumentals he wanted in two days, and he had 28 days left to entry many extra instrumentals. “It is so cheap. If I was still a hustling artist, like in my early days, it would have helped me create multiple albums in a few days,” he mentioned. 

Aside from SoundRaw, Nkasi additionally used AI instruments corresponding to voice synthesizers, ChatGPT, and the picture generator Midjourney AI for the album. ChatGPT wrote the lyrics and steered prompts for different AI instruments, such because the picture generator Midjourney, which was used to create the quilt of the album. 

“I paid for a few other tools that I eventually didn’t use. Of all the text-to-speech AI tools he got, he preferred and used ElevenLabs, as it was the only one that produced a singing voice that truly sounded human. “The AI voice often momentarily paused to take a breath in. There is no reason AI should be breathing, but it was trying to convince me that the voice is human,” Nkasi elaborated.

Read additionally: Fraudsters don’t suppose AI can do the job but

About the album

The album started with an AI-generated voice talking in an American accent and telling a narrative lasting practically two minutes. It delved into the story of a personality named Leo, pushed by a need to create a brand new sound able to uniting individuals from numerous cultures and backgrounds. Leo’s journey took him to a mystical realm often called Infinite Echoes, which had the important thing to a rhythm that connects numerous people and their tales. This narrative of his odyssey resurfaced twice extra all through the album, titled Interlude I and Interlude II. These interludes offered perception into the rationale behind the various genres (hip hop, dance corridor, and afrobeat) in the album.

For occasion, the mellow-paced and poetic monitor “Love Tempo” seamlessly transitioned into the gospel-infused music “God’s Whispers I.” Following the primary interlude, “Dream Chaser” explored the pursuit of materialistic needs, main into “Ifuu (Men Di Bad),” a charming composition enriched with important parts of Igbo lyrics. Once one other interlude concluded, a reggae-like piece titled “Find You,” with lyrics in the language of Patois, took the highlight, concluding with yet one more gospel music, “God’s Whisper II.” 

It could be very difficult for anybody to guess that the sounds, lyrics, and singing are largely a product of AI. Nkasi deemed it vital to immediate the AI to do issues that he would usually do as an artiste.  Songs on the album like “Find You”, and “Ifuu (Men Di Bad)” included lyrics in Nigerian languages like Pidgin, Patois, and Igbo. “What I found very interesting was that we prompted AI to write in Nigerian pidgin, and the results were impressive. On a song like Ifuu that had Igbo; Igbo was AI-generated. The rap verse that refers to palm wine and chilling, was ChatGPT, as well as the Igbo wordings,” he mentioned.

“This time around, I sort of reversed the process. I had the idea for the chorus and wordings for the chorus only, so I typed in a prompt to ChatGPT telling it the chorus and asking it to write out the rest of the song with patois and pidgin language in-between,” Nkasi added. He says that the best way it’s on the music is the best way ChatGPT produced it, and all he needed to do was re-enact it and convey it to life with the opposite AI instruments.

Aside from Ifuu and Dream Chaser, which had important human effort, Nkasi says that each different component was fully created with AI.  “So for the most part, human contribution came from melody syncing because ChatGPT would write the song but because there was no melody to it, I had to figure out the singing melody that would flow best with the beat that AI gave us.” In some circumstances, we wanted to shorten some sentences in the lyrics and lengthen a number of others.

Hitches alongside the best way

Nkasi felt a bit handicapped by the dearth of interoperability of the completely different software program he used to supply the album. “The unique challenge we are facing is that the different AI tools we use to make music do not communicate with each other. If ChatGPT is writing, it has no idea what the beat sounds like. And when we are trying to figure out the melody and create the voice, the AI needs to understand the lyrics and the beats.”  This is why the melody needed to be determined by him (the human). “We took the lyrics we had, came up with melody ideas, trained the AI to sing it with the generated voice, and put them together.”

However, contemplating how rapidly AI improvements are leveling out bottlenecks, Nkasi doesn’t suppose this downside will stay unsolved for too lengthy.  At some level in time, somebody might work out the best way to make these distinct processes join with a plug-and-play technique.

An AI music artist

“A friend of mine, popular gospel artist Limoblaze, argued that it was impossible for AI to write a gospel song people could be moved by. All I had to do was prompt the AI tools to create something that referenced faith and one’s commitment to God. This resulted in God’s Whispers 2, one of the two gospel songs on the album,” Nkasi disclosed.

In God’s Whispers II, Nkasi options an artist recognized as Mya Blue, however she isn’t an actual individual. The different two featured artists, Nnamdi Agbo and David Wonder, are people, however Mya Blue is a female AI-generated voice. Nkasi proved his level, as that music appeared like another gospel music.  He mentioned, “We played the chords, used ChatPGT to write the lyrics, and a completely different AI tool to generate the voice [for Mya Blue]. She doesn’t exist but as an AI.”

The music begins with Nkasobi and Mya Blue doing a duet, and Mya Blue, singing and softly respiration like it’s human, continues with a solo for the final 2 minutes. Its voice jogs my memory of how actress Vannessa Hudgens sang in Disney’s High School Musical.  “I did this for the people who want to convince themselves that they can tell it is AI,” Nkasi added. 

The clampdown on AI

Last month, the Financial Times reported that Universal Music Group directed streaming platforms, together with Spotify and Apple, to dam synthetic intelligence companies from scraping melodies and lyrics from their copyrighted songs.

However, Nkasi thinks that AI is inevitable as there are such a lot of gamers. He says it should profit the typical artiste. “The problem is that the AI can invalidate the huge catalogue that big labels have spent billions of dollars amassing and putting together. For the average artist, it democratises access and makes things easier,” he concludes.