• Slider Image
  • Slider Image
  • Slider Image

The Hybrid Intelligence Work: Can a Layer of Creative Human Intelligence Make Your AI-Generated Work Protectable?

Sticky By Artificial Intelligence, Copyrights, Musical Compositions, Musical Technologies

Creative works, whether written, designed, or musical – are not copyrightable if generated through artificial intelligence (AI).  This is because only human authors are eligible for copyright protection (remember the monkey selfie?). 

What can you do, then, if you feed prompts into a large language model (LLM) like ChatGPT to generate an original story, graphic design, song, or other kind of work?  There may be ways to obtain some remuneration from the work by making it available for download on third-party platforms.  However, as a work with no human author, and therefore no copyright protection, there is nothing to prevent others from copying, displaying, re-distributing, or otherwise exploiting the work for their own personal gain.

Unless there are major changes to the law of copyright, we have to forget about protecting the AI-generated work as is.  But what if you were to make a small, but substantial, change or addition to the work spit out by the LLM?

A creative work that uses a prior work as its raw material and contains changes to that work or adds a sufficient amount of new material is called a derivative work.  If the original work is copyrighted, the author seeking to create a derivative work would need permission from the owner of the copyright in the original work to add to or alter it. 

However, derivative works also can be created from works that are in the public domain.  The most common example of a public domain work is a work whose copyright has expired.  Now that we have works created by AI (and the occasional non-human primate), I think it’s accurate to categorize them as public domain works.

In other words, works generated by AI don’t enter the public domain after a lifetime of copyright protection; they are born into the public domain.

Oh, and here’s the nice thing about a derivative work:  you can get a copyright on it.

But the copyright protects only the additions or changes to the original work, not the elements of the original work.  So, how could you copyright an AI-generated work, and why would you do it?

First, you feed the prompts you want into the LLM – ChatGPT, or whichever one you use – and the AI spits out a work.  Let’s say you’re a composer and the work is a song.  But at this point, the song is merely a public domain work exploitable by anyone.

You’d need to change or add to that song to make it a protectable derivative work of the AI’s original song.  So you compose a new part for the song – vocals, a countermelody, a bridge, and/or a bass line – and using any recording software, you add that new part to the AI-generated song.  Assuming some originality to your new part(s), you now have a hybrid track  – the original AI song plus your new part(s) – eligible for copyright protection as a derivative work.

Let’s call this new kind of derivative work – a hybrid work partially generated by artificial intelligence and partially created by human intelligence – a Hybrid Intelligence Work (HIW).

You file a copyright application with the U.S. Copyright Office and upload the hybrid track to the Copyright Office website with the application.  The application requires that you specify the pre-existing material and the new material, so you provide that information, and you should be able to get a copyright registration for your HIW.

Now, when you release the track yourself and/or via a music streaming service, if someone copies the file, tries to distribute it, etc., that would infringe your copyright in the derivative work. 

Let’s go one step further to illustrate the significance of the HIW.  Maybe you label your track to indicate that it was generated, at least in part, by AI.  There may be some people out there savvy enough to know  that AI-generated works are public domain (and as time goes on and more AI-generated works are distributed, this should slowly become common knowledge).

Even someone armed with this knowledge would have a tough time legally exploiting your HIW because the track consists of an inseparable (or very difficult to separate) amalgam of AI-generated public domain elements and human-created copyrighted elements.  That, in effect, protects your hybrid track from copying because the only way someone can feasibly copy it is to take the whole track, which would constitute infringement of your copyright in the HIW.

I understand that it is possible to isolate parts of a recording, but typically you need access to the original multi-track recording or sophisticated software to do it.  Even if someone had the software and wanted to separate out your copyrighted elements, it might not be obvious which elements of the mix those are.

So, while you can’t currently copyright your AI-generated work, adding a dash of creative human intelligence could result in a copyrightable Hybrid Intelligence Work.

Anthropic Settles Claude Copyright Case

By Artificial Intelligence, Copyrights

Anthropic has reached a settlement with the authors whose books it used for training its AI assistant, Claude. The AI developer has agreed to pay the authors $1.5 billion, according to IPLaw360.

The settlement follows a court ruling in June, which I discussed in a previous post. In that opinion, the court found that training AI is fair use, but copyrighted training materials must be purchased, and maintaining a permanent library of the copyrighted works is questionable.

The court’s central finding was that “the purpose and character of using copyrighted works to train LLMs to generate new text [is] quintessentially transformative.” The court also found that the rounds of copying of legitimately purchased books were reasonably necessary for the transformative training use, and the training did not and will not displace the market for copies of the authors’ copyrighted works.

But obtaining pirated copies, as Anthropic did for some of the books, was not a fair use and constituted copyright infringement. The court also held that the separate use of copying the pirated works and building a central library of them was not a fair use.

So this case will not go to trial, with Anthropic electing to pay a sum certain rather than roll the dice on damages. But the court’s June opinion remains a guide for AI developers, i.e., certain training uses may qualify as fair use, but they need to be careful how they acquire and store the copyrighted training materials.

The Reggaeton Riddim Case: Give the Drummer Some!

By Copyrights, Musical Instruments

I spoke with Ivan Moreno over at IPLaw360 this week about the copyright litigation that could have a major impact on reggaeton, the popular modern blend of hip-hop, Latin, and Caribbean music. Here’s the link to Ivan’s article.

In this huge lawsuit consolidating over 50 cases, Cleveland “Clevie” Browne and other plaintiffs sued more than 160 defendants for copyright infringement. Plaintiffs claim to be the original creators of he distinctive dembow rhythm that’s become a signature of reggaeton music.

They allege that some 1,800 reggaeton songs infringe Jamaican producers Steely & Clevie’s 1989 hit “Fish Market” because they made unauthorized use (either by sampling or copying) of the song’s “riddim,” which is named after Shabba Ranks’ 1990 single “Dem Bow,”

Most of the reporting focuses on the potential impact to the reggaeton genre and whether a “beat,” and more particularly, a “genre-defining” beat is or should be protectable by copyright. But to me that framing is something of a red herring that obscures a fundamental truth about copyright.

Nobody would question that a performance of an original musical composition on a melodic and harmonic instrument like guitar or piano is protectable by copyright. I also think no one would doubt that an original solo recording on a single-note instrument (e.g., sax, trumpet, flute, trombone) can be protected by copyright. 

But when it comes to drums or percussion, where the harmonic and melodic elements are stripped away, our assumptions change. Many people run into a a psychological hurdle acknowledging the protectability, or even existence, of a musical composition when melody is absent from music. 

But it should be an absolutely uncontroversial proposition that a rhythm-only musical composition or recording consisting of exclusively or mostly drums and/or percussion is protectable by copyright, so long as the work is original and has the requisite modicum of creativity. With the harmonic and melodic instruments, each is doing what that instrument can do; that is the same for the drum kit and percussion. 

For starters, the drum kit includes several different “instruments”: bass drum, snare, cymbals, hi-hat, toms, maybe cowbell. Percussion instruments offer endless variation, e.g. tympani, timbale, tambourine, triangle, claves, etc. And, of course, the rhythmic patterns the drummer/percussionist can play also are infinitely varied, with the individual pieces playing different rhythms (usually at the same time), and different kinds of sounds, timbres, etc. 

Plaintiffs try to make this case and even included a score in their summary judgment motion to show how complex the “beat” is in Fish Market:

The top half of the score shows the drum and percussion parts. Seven pieces go into creating the Fish Market “riddim”, and, aside from the hi hat and kick drum, each of the other parts is different.

This demonstrates the complexity of the “riddim” as the high and low timbale, tambourine, synth tom, hi-hat, snare, and kick drum are all playing together in particular way. Indeed, this begins to look like a musical composition, not simply a “beat”  (Of course, calling it a “beat” exacerbates the above-mentioned psychological hurdle).

Whether the drum or percussion composition (the “beat”) goes on to define a genre is irrelevant to the infringement analysis.  If the work is sufficiently creative and original to be protected by copyright and others copy it, that’s infringement, whether it’s a flash in the pan that immediately fades away or subsequent artists make it hugely popular and it comes to define a genre.

I said to Ivan, even though there’s no legal need, it would be nice for a court decision to acknowledge the originality and creativity of a drum/percussion composition. I have many drummer friends and colleagues, and I’m all for increased recognition of their creativity. As I often say during a session or gig, give the drummer some!

Buy, Train, Delete, Repeat: A Generative AI Training Formula Emerges from the Anthropic Decision

By Artificial Intelligence, Copyrights, Musical Technologies

We have what appears to be the first major court ruling on using copyrighted works to train generative AI, and the bottom line is: training AI is fair use because generative AI is “spectacularly transformative,” but copyrighted training materials must be purchased, and maintaining a permanent library of the copyrighted works is questionable.

At issue in this case was Anthropic’s training of its AI assistant, Claude. The company used millions of copyrighted works, some legitimately purchased, some downloaded from pirate websites.

Three authors whose books were used by Anthropic sued for copyright infringement, and Anthropic moved for summary judgment that the training is not infringement but instead a protected fair use.

The court’s opinion, by Judge William Alsup of the U.S. District Court for the Northern District of California, is a detailed and logical analysis of copyright fair use as applied to sourcing, digitizing, and storing copyrighted works as well as, of course, using the works to train generative AI. The court necessarily (and helpfully) breaks out and separately analyzes the different acquisitions and “uses” of the books as well as the different stages of copying the books during the whole process.

Anthropic acquired the millions of books it needed for training purpose in two ways: downloads from pirate libraries and legitimate purchases of mostly used books. The physical books were scanned or digitized into the company’s central library, and all the books stored there. Significantly, not all of the books were used for training and the central library was kept permanently as a general-purpose collection of books even after the training was completed.

During the course of the process, the books were copied in several different ways: first, the books were copied from the central library to create working copies for the training set; second, each work was cleaned to remove lower-value or repeating text, resulting in clean copies; third, each cleaned copy was translated into a “tokenized” copy for training; fourth, the tokenized books were repeatedly copied during the training.

As to the “uses” of the books, for its fair use analysis, the court found two. The first way in which Anthropic used the books was to build a huge central library of content. The second is the obvious one, i.e., to train the large language models (LLMs) for its AI assistant.

The court went through the four statutory fair use factors for each use, discussing how each factor applied to the training copies, the purchased copies, and the pirated copies for both uses.

The court’s central finding is that “the purpose and character of using copyrighted works to train LLMs to generate new text [is] quintessentially transformative.” This is because such training is not to “race ahead and replicate or supplant” the copyrighted works but instead to “turn a hard corner and create something different.”

The court also found that the rounds of copying of legitimately purchased books were reasonably necessary for the transformative training use, and the training did not and will not displace the market for copies of the authors’ copyrighted works. Thus, the copies used to train Anthropic’s LLMs are not infringing, and the training is a fair use.

As to the purchased copies Anthropic converted from print to digital, the court reasoned that such a format change did not add any new copies but rather eased storage and searchability without damaging the copyright owners’ interests. The format change also did not relate to a right reserved to authors of copyrighted works under the Copyright Act. Thus, the change from print to digital was a fair use.

But obtaining pirated copies of the books was not a fair use and constituted copyright infringement. The court said “the person who copies the []book from a pirate site has infringed already, full stop.” The court also held that the separate use of copying the pirated works and building a central library of them was not a fair use.

Finally, as to copies made from the central library copies but not used for training and copies retained even when no longer serving as sources for training copies, the court reserved judgment and denied Anthropic’s motion for summary judgment of fair use.

With this decision, the contours of copyright fair use law as applied to training generative AI are taking shape. The big takeaway is that developing training copies of copyrighted works and using them for training are protected fair uses.

However, to enjoy this protection, those training AI tools should adhere to a couple of ground rules: first, obtain the copyrighted works legitimately, i.e., buy the books; don’t copy pirated versions. Second, to the extent copies of the copyrighted works are not used for training or once training is complete, one might need to delete the works rather than maintain them permanently. That is, future trainings might require a new round of purchases.

For the generative AI training formula, then, a useful mantra might be: buy, train, delete, repeat.

Unsigned, Sealed, Delivered: Are Copyright Certificates Valid Without a Register?

By Copyrights

If you recently received a certificate of copyright registration from the U.S. Copyright Office or expect to in the near future, be vigilant because they may be unsigned.

Along with other chaos and uncertainties caused by the Trump administration, the Copyright Office is currently without a Director and Register of Copyrights because Shira Perlmutter was fired. Thus, currently there is no individual filling the appropriate role of signor of certificates.

Although the Copyright Act does not expressly require that a certificate of registration be signed, there is, unfortunately, a question as to whether an unsigned certificate of registration is valid.

This is because Section 410(a) of the Copyright Act says it is the Register of Copyrights who “determines” whether the deposited work is copyrightable and registers the work and issues the certificate of registration:

(a) When, after examination, the Register of Copyrights determines that, in accordance with the provisions of this title, the material deposited constitutes copyrightable subject matter and that the other legal and formal requirements of this title have been met, the Register shall register the claim and issue to the applicant a certificate of registration under the seal of the Copyright Office.

Absent a Register to make the appropriate determinations of copyrightability and issue certificates of registration, it is unclear whether an issued certificate would comply with the statute.

This point may need to be litigated, but we can only hope that the administration gets a new Director and Register in there quickly.

String Theory: Larsen’s New String Production Patent

By Music Gear and Accessories, Musical Instruments, Musical Technologies, Patents

Larsen strings are considered to be quite good, especially for viola and cello. Musicians like their warm, bright tone, strong projection, and good responsiveness.

The Danish company, Larsen Strings, recently obtained a U.S. patent for its string fabrication apparatus and process. Bearing the punchy title “Method for fabicating a string, in particular a string for a bowed musical instrument, and an apparatus for carrying out the same,” U.S. Patent No. 12,281,437 (437 Patent) issued April 22, 2025.

A string for a bowed musical instrument such as a violin or cello typically consists of a core material with one or more layers of winding materials.

According to the ‘437 Patent, the compactness of the core and the winding layers is an important property of a string. If a string lacks compactness, the core and the winding layers are not sufficiently interlocked, and when the string is under tension, the layers may shift relative to one another, leading to poorer string response. The goal of the patented fabrication method is to produce strings with increased compactness.

The ‘437 Patent is directed to a method and apparatus for fabricating a string (110) where a compactness increasing module (120) applies both a friction force and a compression force to a winding strand (4) at the spinning point (7) as the strand is wound around a cylindrical core (3).

String fabricating apparatus (100) produces a string (110) having a core (3) with at least one winding strand (4) helically wound around the core. The apparatus (100) rotates the core (3) and winds the strand (4) around the core as the core rotates.

The compactness increasing module (120) contacts the winding strand (4) at a spinning point (7) when the strand is wound onto the core (3) so a friction force and a compression force are introduced at the spinning point.

The friction force increases the friction between the compactness increasing module (120) and the winding strand (4), and the compression force compresses the winding strand (4) and the core (3).

According to their website, Larsen strings made by the patented process produce a consistently perfect bow response and an accurately calibrated sound.

Undercover: AI-Generated Songs can be Cover Songs or Infringing Copies

By Artificial Intelligence, Copyrights, Musical Technologies

As discussed in previous posts (e.g., here, here, and here), last year the Recording Industry of America (RIAA) sued two major AI music platforms alleging that their use of copyrighted musical works to train their AI software infringes the copyrights held by several record companies.

As noted in the court filings, the RIAA does not allege that the Suno and Udio user outputs – the AI-generated musical works themselves – are infringing.

What would alleged copyright infringement by an AI-generated song look like? Its probably just a matter of time before an AI-generated musical work becomes a commercially successful release, and an artist rightsholder thinks the machine song too similar to the human song.

The starting point is that AI music generators typically are trained on copyrighted material, so entering certain kinds of prompts could yield an outputted song that closely resembles one or more of the copyrighted musical works fed into the AI software. The Complaints filed by the RIAA illustrate this phenomenon (not to show infringing outputs, but to demonstrate that the training materials included copyrighted songs):

The above example demonstrates that an AI-generated song – Sunshine Melody – could quite reasonably infringe the musical composition copyright in the famous Temptations song, My Girl.

But in a case like this would the output necessarily be an infringing work? That’s definitely possible. But another possibility is that the outputted song is a cover of the Temptations song.

Why is this distinction important? Because you don’t need permission from the owner of the copyright in a musical composition to create and digitally distribute a cover version.

Cover songs are governed by a compulsory license regime set out in the Copyright Act. The artist recording the cover has to jump through certain statutory hoops – including providing notice to the copyright owner and paying the royalty fees based on the statutory rate – but the copyright owner cannot prevent covers from being created and digitally distributed.

We all know what a cover song is, but it’s worth stating the definition here: it is a new recording of an existing sound recording, performed by someone other than the original artist.

A cover is faithful to the original in important ways (title, melody, lyrics, maybe structure and chord changes), but often the artist creating the cover song will try to bring her own style and sound to it. How might these conventions apply to an AI-generated musical work?

Let’s get into the weeds using the My Girl example. If the user who entered the prompt to generate the new song depicted above and titled it “My Girl” it seems clear that it would be a cover. It has the same title, the same lyrics, the same chord changes, and a very similar vocal melody line.

Even if there are some significant differences, e.g., in vocal sounds, instrumentation, tempo, style, feel, it would still be a cover song. After all, the cover artist can bring creativity and a different aesthetic while remaining faithful to the original.

But what if the AI user released the new song under the title “Sunshine Melody” instead? That probably would constitute infringement of the musical composition copyright in “My Girl”. At least, it makes it more difficult to argue that the intent was to create a cover.

What if the new song has the same chords, same structure, and same vocal melody but new lyrics? That also is likely to be copyright infringement; that song feels like a new work where the artist engaged in wholesale copying of the original (note, I’m putting aside for this discussion parody and related analysis, e.g., Weird Al etc.).

So it’s complicated. Of course, these considerations are relevant in the context of a traditionally created song as well as that of an AI-generated work, but in the brave new world of AI-generated music existing gray areas can become even grayer. And because copyright law applies equally to the realm of AI music, these are issues we need to think about.

Photographer Asks Supreme Court to Reject the Server Test for Copyright Infringement

By Copyrights

The Court of Appeals for the Ninth Circuit out on the west coast does not always move in step with the other federal appeals courts. The Server Test for copyright infringement is a timely example, and the issue is now on the doorstep of the Supreme Court.

The Copyright Act provides that the owner of the copyright in a pictorial work has the exclusive right to control how the work is publicly displayed. That is in addition to the other rights in the so-called “bundle of rights” such as the rights of reproduction and distribution.

In a case called Perfect 10, Inc. v. Amazon.com, Inc., the Ninth Circuit held the following, making a distinction based on the technological means by which a copyrighted image is displayed on a website:

  • If a website publisher displays an image by filling “a computer screen with a copy of the photographic image fixed in the computer’s memory”, that is copyright infringement.
  • However, if a website publisher displays an image by “embedding” a link to it, i.e., the website’s backend HTML code “gives the address of the image to the user’s browser” and the browser “interacts with the [third-party] computer that stores” the code representing the image, that is not copyright infringement.

The Perfect 10 decision concluded that, because the computer code of the image remains on a third party’s server and is not fixed in the memory of the website operator’s computer, the image is not “displayed” on the operator’s website per the Copyright Act, and therefore there is no infringement of the display right.

Notably, other Circuits, including the Second, Fifth, and Tenth, have rejected the Server Test.

Elliot McGucken is a nature and landscape photographer. Valnet owns and operates a travel website, www.thetravel.com. McGucken alleged that Valnet displayed 36 of his photographs taken from his Instagram account in online articles about travel destinations.

The district court and the Ninth Circuit Court of Appeals ruled against McGucken under the Server Test of Perfect 10 because Valnet’s website contained only embedded references to McGucken’s Instagram posts.

In late March, after losing in the Ninth Circuit, McGucken asked the Supreme Court to hear the case. In his Petition for Certiorari, the question presented is:

Whether the exclusive right to publicly display a copyrighted work . . . is infringed when a website operator publicly shows a copyrighted work without authorization, regardless of the technological process used to show the work.

More specifically stated in the Petition, is the display right infringed “when a website employs technical means to display a copyright image without storing that image on its own servers.”

McGucken’s main argument is that the Server Test is inconsistent with the text of the Copyright Act. The Act separately confers both the exclusive right of reproduction and the exclusive right of display.

However, under the Server Test, a website infringes the display right only if it first makes a reproduction of an image and stores it on its own server. Thus, the Server Test collapses the display right into the reproduction right, rendering them no longer separate rights:

Under the Server Test, one cannot display . . . without first reproducing, which makes the display right mere surplusage.

A circuit split as there is here makes it more likely that the Supreme Court will grant cert and take up the case. If that happens, I suspect they will slap down the Ninth Circuit and eliminate the Server Test. To me it seems like legislating – poorly – from the bench.

Oh, the Humanity: For Copyright, Machines Need Not Apply

By Artificial Intelligence, Copyrights

The question of whether a non-human author can own a copyright has been the subject of much discussion, including Copyright Office opinions and circulars (see my previous post on a recent Office report on AI and copyright).

The most famous case – one that captured the public imagination – is the case of the monkey selfie, in which a photographer applied for copyright registration for a photo taken by a crested macaque using a camera the photographer had set up in the forests of Indonesia.

The U.S. Copyright Office ruled that only works created by a human being can be copyrighted; the Office “will not register works produced by nature, animals, or plants.” Notably, the list does not include software or machines.

Though the human creation requirement didn’t seem promising for authors working with artificial intelligence, it seemed like maybe the door was open just a crack.

Now that door has definitively closed. In a recent decision, the U.S. Court of Appeals for the District of Columbia Circuit held that a non-human machine cannot be an author under the Copyright Act.

The lawsuit, Thaler v. Perlmutter, was brought by Dr. Stephen Thaler, a computer scientist who created artwork using AI. He filed a copyright application for the work (below, called “A Recent Entrance to Paradise”), listing the “Author” as the “Creativity Machine” and himself as the “Copyright Claimant,” i.e., the owner of the work.

After the Copyright Office denied Dr. Thaler’s application on the ground that “a human being did not create the work,” he went through the Office appeal process, then to federal district court, and up to the Court of Appeals.

In a thorough opinion, the appeals court affirmed that the Copyright Office appropriately denied Dr. Thaler’s application because “the Copyright Act requires all work to be authored in the first instance by a human being.”

The heart of the opinion is the court’s analysis of the Copyright Act’s language relating to “authors” and “machines.” The court runs through a series of statutory provisions that:

both identify authors as human beings and define “machines” as tools used by humans in the creative process rather than as creators themselves. Because many of the Copyright Act’s provisions make sense only if an author is a human being, the best reading of the Copyright Act is that human authorship is required for registration.

First, the court notes that the Copyright Act’s ownership provision is premised on the author’s legal capacity to own property, and a machine cannot own property, so it cannot be an author under the statute.

Second, the Copyright Act ties the term of copyright protection to the author’s lifespan, expiring 70 years after the author’s death. Machines do not have “lives” and their operability period is not measured like a human life, so they cannot be authors under the statute.

Third, the inheritance provision of the statute provides for passage of copyright to a surviving spouse or children, and machines, of course, do not have spouses or other heirs.

Fourth, a copyright transfer requires a signature, and machines lack signatures and the legal capacity to provide authenticating signatures. Fifth, authors of unpublished works are protected regardless of “nationality or domicile,” and machines do not have domiciles or nationalities. Sixth, for a joint work, the authors must have the intention to collaborate, while machines do not have intentions.

Finally, the court noted that every time the Copyright Act discusses machines, the context indicates that machines are not authors, but are merely tools to assist authors. For instance, a “computer program” is defined as “a set of statements or instructions to be used directly or indirectly” to “bring about a certain result.” The statute defines the term “machine” the same as “device” and “process,” i.e., as mechanisms that assist authors, rather than authors themselves.

The court summed up:

All of these statutory provisions collectively identify an “author as a human being. Machines do not have property, traditional human lifespans, family members, domiciles, nationalities, mentes reae, or signatures. By contrast, reading the Copyright Act to require human authorship comports with the statute’s text, structure, and design because humans have all the attributes the Copyright Act treats authors as possessing.

Therefore, where copyright is concerned, only humans need apply: “the current Copyright Act’s text, taken as a whole, is best read as making humanity a necessary condition for authorship under the Copyright Act.

Amid this flurry of legal activity around humans, non-humans, and copyright, the framework laid out in my maiden post – adding some human creativity to an AI-generated work – remains the only viable path to copyright protection for artists using AI.

Trippy: Casio’s Patent-Pending Dimension Tripper Allows Pedal Control with Guitar Strap

By Music Gear and Accessories, Musical Instruments, Musical Technologies, Patents

If you’re a guitarist (or play with a guitarist) you know the deal. Before counting off a musical number, as the band reaches a transition in a tune, or just before a guitar solo starts, the eyes shift to the board on the floor, search for the pedal for the desired sound or effect, and the front of the shoe clicks on the button of the chosen pedal.

That common sequence of events may soon become a thing of the past. Casio’s Dimension Tripper allows you to control your guitar pedals with your strap. It may be the only wireless expression controller available that can be operated with the guitar strap.

Casio’s patent application for the system, published February 13th, provides a peek behind the curtain at how the Dimension Tripper works.

U.S. Patent Application Publication No. 2025/0054471 is entitled “Electronic apparatus for musical instruments” and directed to a playing system (1) where a change in tension applied to the strap (210) sends a signal that results in the guitar playing a particular sound or effect.

Playing system (1) includes an effector apparatus (100), an electric guitar (200), and a guitar amplifier (300). The electronic apparatus (10) is located between the guitar strap (210) and the electric guitar (200).

The effector apparatus (100) has an effect unit (120) connected to an electrical signal input unit (110), an electrical signal output unit (130), and a control signal input unit (140). Electrical signals (sound signals) travel from the electric guitar (200) to the electrical signal input unit (110).

The electrical signal output unit (130) is connected to the guitar amplifier (300). The electrical signal output from the electrical signal output unit (130) is eventually emitted from the guitar amplifier (300) in the form of sound.

When the guitar player presses the electric guitar (200) downward to apply a tension to the strap (210), a rotational shaft (41) of a rotation detection device (40) rotates in accordance with the tension applied to the strap (210).  A rotation amount of this rotational shaft (41) is detected by the rotation detection device (40) and is sent to the receiver (150) as a control signal so the effector apparatus (100) can be controlled.

The effect unit (120) applies various types of effects e.g., wow, distortion, etc., to the electrical signal input from the electrical signal input unit (110), and the resulting electrical signal is output by way of the electrical signal output unit (130).

Thus, the Dimension Tripper is essentially an expression pedal. But unlike other forms of expression controllers where the guitarist rocks a pedal back and forth with his foot or slides a fader back and forth with her hand, the guitarist pulls down on the guitar itself to change the tension in the strap.

You can see how it works in this video by engadget from the NAMM show.

TapTunes Provides New Physical Media for Music: A Conversation with Gino Gavoni

By Artificial Intelligence, Musical Technologies

78s, 45s, LPs, 8-tracks, cassettes, CDs. The list ends there. There are digital music files and streaming of course. What I mean is the list of physical media for music ends with CDs.

Not if Gino Gavoni has anything to say about it.

Gavoni has invented a new physical medium artists can use to distribute their music. Using Near Field Communication (NFC) technology, the Tap Card is a small device that provides an NFC-embedded connection to audio media players.

I’ll get into more details on the Tap Card shortly, but let’s start at the beginning, which I was fortunate to be able to do in my interview with Mr. Gavoni.

Gavoni has been a musician, performer, and composer nearly all of his life, starting with the guitar in high school. At nineteen, his guitar teacher encouraged him to audition for the big band of legendary arranger Nelson Riddle.

At the audition the young Gavoni told Riddle that his arrangement of “Spanish Eyes” contained some mistakes. When he pointed out the errors, Riddle took a pencil from behind his ear, scribbled in the corrections, and told Gavoni “you’ve got the gig!”

Later Gavoni spent a number of years in the pro audio department at Paragon Music in Tampa, Florida, becoming an expert in microphones on the advice of Paragon’s owner, Dick Rumore. Rumore told him “if you learn about microphones, you’ll never go hungry” because it’s the one thing every musician needs, no matter what they play.

Gavoni turned to advertising and marketing after that, in 1999 developing a successful coupon book called “Coupons a la Carte”.

From there, Gavoni combined his love of music and his promotional nous to found the Brand O’Guitar company. With his wife, Lisa, Gavoni’s new company provided custom-branded musical instruments.

The business took off quickly. The first phone call they received was from Coca Cola’s advertising representative who had seen Gavoni’s mockup of a Coca Cola branded guitar on the Brand O’Guitar website. The ad rep loved it and asked how they could buy one.

Which brings us to TapTunes. Gavoni got the idea after watching The Playlist, a movie about Spotify. Struck by how streaming has dented the revenue of most artists, he started thinking about ways to help musicians make money.

It occurred to him that a new physical medium, picking up where CDs left off, might be the answer. The Tap Card was born.

Gavoni started TapTunes, which sells the cards to musical artists. When an artist orders a starter package ($99), she gets 25 Tap Cards, a code with a link, and instructions on how to set up an album on a server. With their music now on something tangible, the artist can sell the Tap Cards to her fans or distribute them as promotional items.

The patent-pending cards are small devices, each having a QR code. The music fan either scans the code or taps his phone with card, the media player goes to the server, identifies and grabs the data, and the artist’s album is loaded onto the phone’s media player via an NFC-embedded connection.

In addition to artists, Gavoni pitched Tap Cards to some big brands, which became clients. Spirit makers such as Tito’s Vodka, for example, with big presences at music festivals, were interested in buying branded cards of unique music for promotional purposes.

Then he turned to AI. Gavoni has been using it for a while, and now regularly uses it as an aid in composing music. “AI is the best songwriting partner I have ever had” he told me.

At taptunes.ai, Gavoni makes the pitch for Tap Cards to artists using AI to generate their music:

In the evolving landscape of music, AI creators face unique challenges in monetizing and distributing their work.

TapTunes™ NFC Music Cards offer a revolutionary solution. These innovative cards work seamlessly with every smartphone, putting your music directly into the hands of nearly every potential listener. With TapTunes™, you become your own publisher and distributor, retaining full control of your content and earning potential.

A compelling pitch for the future of music distribution from a music technologist to all the music technologists out there.