"You'll never read a manual again": How AI can help you learn to use your gear and instruments
Mylar Melodies explains how Google's Notebook LM can digest product manuals, help you get to grips with your gear, and... launch your podcasting career?

"Read the fucking manual." It's an instruction given so frequently to music-makers struggling to figure out their gear that it's even spawned its own acronym: RTFM. It's given for good reason, too: nine times out of ten, the answer to your question can be found right there in the manual.
Reading the fucking manual isn't always as simple as it sounds, though. Depending on the piece of gear you're learning, that manual can take the form of anything from a mere pamphlet to a dense and forbidding tome, littered with indecipherable diagrams.
Both of these present unique challenges; not enough information, and you're left scratching your head; too much, and you're sifting through pages and pages of irrelevant content only to answer a simple query. Even the most enterprising gear nerds amongst us may not have the time or the focus to fully digest a 300-page manual. Not to mention that, for those with conditions such ADHD and dyslexia, trawling through such a document is an even more onerous task.
YouTuber, podcaster and electronic musician Mylar Melodies has a solution. In his latest video, the self-described "synthesizer televangelist" explains how to transform Google's AI-powered research tool Notebook LM into an "indescribably helpful" assistant that can instantly answer any question you might have about the gear you're learning to use.
In the video, Mylar Melodies uploads the product manual for the Torso T1 sequencer in PDF format to Notebook LM. After a thirty-second pause, Notebook LM has digested the 300-page manual in its entirety and is now ready to answer any questions on the T1 that are thrown at it.
"How do I set up MIDI channels on the Torso T1?", Mylar asks his newborn sequencer-savvy chatbot. Without missing a beat, Notebook LM returns a thorough explanation that - crucially - provides citations that reference the manual itself. "How do I create custom user scales?", he asks again. Voila: Notebook LM provides a step-by-step explanation that guides him through the process.
Out of curiosity, we've tried a similar thing with ChatGPT in the past and found that it's far too susceptible to hallucination - the name given to AI's tendency to confidently spout rubbish - to act as a reliable assistant in learning how our kit works. As Mylar explains, his method avoids this problem by working with an AI model that limits its sources to only the documents uploaded by the user - documents that we know are reliably accurate.
Get the MusicRadar Newsletter
Want all the hottest music and gear news, reviews, deals, features and more, direct to your inbox? Sign up here.
We're now able to have a conversation with the manual, in the form of a helpful and informed assistant that can patiently walk us through any aspect of the gear we're trying to learn
Notebook LM seems like a useful tool, then, if not quite mind-blowing in its capabilities. Think again. As Mylar Melodies seemingly discovers halfway through recording the video, Google's supercharged research assistant can not only digest manuals and offer helpful written responses, but it can also generate a 'podcast' - complete with mildly irritating synthesized voices - that casually talks through the contents of the manual as if they were the subject of a real conversation.
Not just any old conversation, however: this is a conversation that you, the listener, can participate in, in real time. Rather than laboriously thumbing through our manuals, or even typing in questions and receiving answers about it, we're now able to have a conversation with the manual, in the form of a helpful and informed assistant that can patiently walk us through any aspect of the gear we're trying to learn. The mind boggles.
AI has understandably earned a spotted reputation in recent years, but here, it seems that Mylar has stumbled on an application for artificial intelligence that is both harmless and immensely useful for music-makers. "I find this shockingly effective, scary and kind of amazing," Mylar says. "This, to me, feels like this is what AI should be used for."
Support Mylar Melodies on Patreon or subscribe to his YouTube channel.
I'm MusicRadar's Tech Editor, working across everything from product news and gear-focused features to artist interviews and tech tutorials. I love electronic music and I'm perpetually fascinated by the tools we use to make it. When I'm not behind my laptop keyboard, you'll probably find me behind a MIDI keyboard, carefully crafting the beginnings of another project that I'll ultimately abandon to the creative graveyard that is my overstuffed hard drive.
You must confirm your public display name before commenting
Please logout and then login again, you will then be prompted to enter your display name.

“The perfect environment for creating iconic drum sounds, which have scope to move beyond their original sonic remit”: GForce Software IconDrum review

"I love that sound - it's the reason I bought a Polymoog": UJAM's Vox Humana faithfully captures the classic Polymoog preset from Gary Numan's Cars