Brilliant Disguise
Unpacking Google's new AI music generator; Project Voice 2023 lineup nears completion; version 1.1 of Conversational AI Industry Landscape Map released
So tell me what I see
when I look in your eyes…
Is that you, baby?
Or just a brilliant disguise?
Bruce Springsteen’s Brilliant Disguise, the first single off Tunnel of Love (which was the follow-up to the mega-album Born In The USA), pulls off a unique feat: it got popular (peaked at #5 on the US singles chart) while being extremely dark.
Written at the time of Springsteen’s divorce from his first marriage, the album itself confronts difficult topics and for that reason, became an enduring favorite of fans.
The music video for Brilliant Disguise does something completely unprecedented in this new digital, synthetic, highly-produced era:
it features a singular scene with the artist himself, sitting at a table, singing in a live track over the recorded backing music.
The raw nature of the video only adds to the raw nature of the song’s writing, leaving us with something deeply human.
Sounds like something we could use more of.
We’ve talked about ChatGPT…a lot.
Conveniently, this past week, we got news of a new generative AI tool:
Google has created an AI music generator that creates music based on your text.
Yep. ChatGPT, but for music this time.
From The Verge’s reporting on the subject:
Google researchers have made an AI that can generate minutes-long musical pieces from text prompts, and can even transform a whistled or hummed melody into other instruments, similar to how systems like DALL-E generate images from written prompts (via TechCrunch). The model is called MusicLM, and while you can’t play around with it for yourself, the company has uploaded a bunch of samples that it produced using the model.
The examples are impressive. There are 30-second snippets of what sound like actual songs created from paragraph-long descriptions that prescribe a genre, vibe, and even specific instruments, as well as five-minute-long pieces generated from one or two words like “melodic techno.” Perhaps my favorite is a demo of “story mode,” where the model is basically given a script to morph between prompts.
“Story Mode” got my attention. (Actually, let’s not pretend, the whole story got my attention - this is a newsletter where every new issue is themed around a song.)
The Verge’s article gave an example of a prompt that a user gave to MusicLM:
electronic song played in a videogame (0:00-0:15)
meditation song played next to a river (0:15-0:30)
fire (0:30-0:45)
fireworks (0:45-0:60)
And, from this text, MusicLM created this:
That music is better than many songs I’ve heard in videogames, or other mediums for that matter, over many years.
I’m surrounded by companies on a daily basis doing amazing work with AI, and even I think this is utterly insane.
Lest we immediately jump to the conclusion this technology will cost people jobs - maybe producers, maybe songwriters, maybe artists - I would direct you to a recent headline I happened to see written about ChatGPT:
A profession that technologists and startups have tried to kill over and over and over again through the years - real estate agents - are celebrating the arrival of AI.
Yep, AI. You know, that thing that was supposed to eliminate all their jobs?
This wasn’t quite the script we expected.
For now.
The great thing about music is that it captures our humanity.
So you can expect AI-created music is going to challenge us.
What if the AI sings about killing people?
Is that the same as Johnny Cash singing about killing people?
And if you think the lawyers are going to stop it, you might have another thing coming.
The unavoidable outcome of all this AI in the media, AI in the news, AI everywhere, is to make us appreciate truly great human artistry.
Such as the music of Bruce Springsteen.
Not a bad outcome.
We are pleased to announce the addition of Sarah Coward (left; of the UK's In The Room) who has brought conversational AI to the Holocaust, to be the closing keynote of Project Voice 2023.
Additionally, Kane Simms of VUX World, the well-known beating heart of conversational AI and digital transformation within contact centers, will be part of Project Voice 2023 as well, giving a talk and MCing the Contact Center 2.0 breakout track of the conference.
The powerhouse program for Project Voice 2023: The World of Conversational AI sits at approximately 75% complete, and can be viewed here.
Version 1.1 of the highly-anticipated Conversational AI Industry Landscape Map, a joint product between Project Voice 2023 and Project Voice Capital Partners Rolling Fund (made available via AngelList), has now been released.
The 100+ MB high-resolution PDF can be downloaded here, at no cost.
Version 1.2 of the Map will be released in conjunction with Project Voice 2023 in late April. If you have suggested additions or revisions, email info@projectvoice.ai.