8Sided Blog

a zine about sound, culture, and the punk rock dream

A DJ’s Spin on Music Recognition Tech

· Leave a Comment

It’s no secret that a popular topic on this blog is audio recognition technology as a means to determine royalty payments from public establishments. I wrote about it yesterday. And I wrote about it previously. And previously.

PRS is testing music recognition technology in bars and nightclubs in the United Kingdom. According to Synchtank’s blog, there’s been an unexpected roadblock:

[AEFM’s Greg Marshall] adds that they are speaking to DJ agents to ensure they are getting on board with the moves by clubs and bodies like PRS, specifically addressing the non-record clause some DJs have in their contracts (to ensure their sets are not bootlegged or what they are playing is listed online). […] “We are just starting the process of talking with the bigger agents to see if they can amend their clause so it has an exception for music recognition technology,” explains Marshall.

I could see privacy (eavesdropping) concerns as a complaint against constant monitoring. But we are now allowing smart speakers/listeners in our homes so ¯_(ツ)_/¯. And I imagine the venue’s sound system is directly fed into the monitors, and not a literal microphone hanging out in the club.

I doubt DJs are worried about PRS recordings their sets or that monitoring reveals all their super-secret hottie tracks. The problem is a clause traditionally appearing in most DJs’ contracts that prevents recordings from happening for those reasons. The venues, understandably, want to adhere to the artist contract.

Is audio monitoring technically a recording? Regardless, this is a wild hiccup in the effort to use music recognition tech to determine accurate venue performance royalty distribution. DJs — add an exception in your contracts!

🔗→ Electronic Money: Dance Music and the Protracted Pursuit of Payment

Dubset’s Major Move

· Leave a Comment

TechCrunch:

Spotify and Apple Music could soon get the legal grey area of music like remixes and DJ sets that today live unofficially on SoundCloud. Sony Music Entertainment today became the first major record label to allow its music to be monetized through unofficial mixes thanks to a deal with rights clearance startup Dubset. That means Sony’s master recordings will be indexed by Dubset, and rights holders will be compensated even if just a tiny one-second snippet of their song is used in a DJ set or remix.

A source tells TechCrunch that Dubset is getting closer to securing deals with the other two major labels Warner and Universal.

If it can lock down all three, remixes and DJ sets featuring almost any music could be legally hosted on the top streaming services instead of being barred or removed for copyright infringement. That might eliminate the differentiation that’s kept struggling SoundCloud afloat. Illegal music uploaded there has sometimes flown under the radar since SoundCloud is protected by Safe Harbor law regarding user generated content. But if it’s legally available on Spotify, Apple Music, and elsewhere, listeners wouldn’t have to go to SoundCloud.


Could we be stepping closer to a mainstream acceptance of remix culture? A future where derivative works are not only allowed but encouraged is a divergent music future, indeed. As previously stated on this blog, if you can clear unauthorized remixes using Dubset, then why not clear samples eventually? We might be entering an era where most music is fair game for creative mutation, and the original artists get paid. How will that work with songs already released, especially the ones that sneakily didn’t clear drum loops or other samples? Should clearance lawyers start looking at new career options?

As far as Apple Music and Spotify go, I really can’t see them opening up their services to user-uploaded content a la SoundCloud. I’m ready to be surprised, but I do think those predictions are off the mark. The Verge gives a clue to where this might be headed for the two big streamers:

DJ mixes have historically proved to be especially difficult for monetized distribution. “The average mix is 62 minutes long and has 22 different songs in it, and those 22 different songs are represented by over 100 different rights holders,” {Dubset CEO Stephen} White tells The Verge. Using Dubset’s technology, a 60-minute mix can be processed in just 15 minutes.

During that 60-minute mix, White says, MixSCAN will fingerprint every three seconds of audio. “We’re using a combination of audio fingerprinting technologies and fairly advanced algorithmic approaches to identify the underlying masters that are being used in a mix or a remix,” he says. Although MixBANK asks DJs themselves to identify the masters, White says this is just to help validate MixSCAN’s results.


Apple’s Beats 1 Radio regularly broadcasts sets by newsworthy artists and celebrities, but the Beats 1 platform still fails to make the news. These DJ events need exposure outside of the ephemeral original broadcast. Wouldn’t it be nice if the sets were recorded and archived, and then available to play on demand via Apple Music? I think that’s what’s happening here. A different sort of license is required to make these DJ sets available on demand, and every song (and, yes, unofficial remix) must be cleared for this type of usage. Theoretically, Dubset’s technology would not only clear the songs in the mix, but it would be able to do so in 15 minutes. A Beats 1 set could be available to stream on Apple Music within thirty minutes of its broadcast. Voilà. And I’d wager Spotify has similar ambitions.


Previously and Previously and Previously

Destroying the Perfect

· 5 Comments

On the Building New Law podcast, Seth Godin had this to say:

We always destroy the perfect before we enable the impossible. For example, sonically CDs are not as good as vinyl, and MP3s are not as good as CDs. But this degradation is necessary to get to the technological point of ‘every song in your pocket,’ and audio quality will someday catch up.


And we’ve seen it before. The eight-track tape: sounded like crap but you could play it in your car. Then came the cassette, also crappy but you could go for a run with a Walkman at your hip. Compact discs eventually improved the quality and kept the mobility. But there’s another level of convenience that no one anticipated, which is the convenience of library and access. This facet was the promise of “every song in your pocket,” and that means it was a step back to move forward, courtesy of relatively lo-fi MP3s.

Are we now at the technological point of ‘catch up’ Godin mentions? For many of us, the bandwidth is now there, and bandwidth has been the primary constraint. Is it time to seriously upgrade our stereo systems for streaming? From BBC News:

Qobuz, along with rivals Tidal and Deezer Elite, offers streaming of “lossless audio” that throws nothing away.

“Is MP3 as interesting as it was ten years ago? Not really, because bandwidth has improved,” says Malcolm Ouzeri, head of marketing at French streaming and download provider Qobuz, founded in 2007. “Now the industry is going towards more quality.”

The highest quality MP3 has a bit-rate of 320kbps, while a hi-res file can go as high as 9,216kbps. Music CDs are transferred at 1,411kbps.


There is also talk of Spotify launching a lossless audio option. Some users report seeing this option in limited test cases. And then there’s the adoption of the LUFS standard by Spotify and other streamers, showing a renewed attention to sound quality. But many of these services make hi-res an add-on option. The rumor is that Spotify’s hi-res audio will be available as part of a more expensive monthly plan, as Tidal currently offers. A Qobuz ‘highest quality’ subscription is presently £349.99 a year.

I’m not sure if hi-res audio will make an impact as long as it’s seen as an add-on for those with extra change to spare. Even the option titles – such as Deezer Elite – make hi-res seem elitist. I don’t know what the additional costs are to the providers, but it will be wonderful to finally enter a world where hi-res audio is a sole and affordable option as bandwidth grows and accelerates. Once we’ve arrived, the only ticket for entry will be our choice of speakers.

Songwriters Getting Paid as the Robots Listen

· 4 Comments

There are a few options for businesses to legally play music on premises, whether that business is, say, a nightclub, restaurant, or hair salon. An in-store music service like Mood Media (formerly Muzak) can supply channels of pre-cleared tunes for a subscription fee. These services are like radio in most cases, as the business won’t be able to choose any particular song that’s played. The business could also just play music by friends and enter into a direct licensing agreement with each songwriter. That would be a huge hassle and dramatically limit the available catalog.

The most popular option is to pay for the compulsory licenses offered by the performance rights organizations – PROs like ASCAP, BMI, and SESAC. These licenses allow the business to play songs represented by each PRO. A few factors determine the fee, such as venue capacity, and the business usually obtains a license from all the PROs. For one thing, it’s a lot of work to determine which PRO represents a song the venue wants to play, and the music customers would like to hear are distributed amongst all the rights organizations. Paying fees to all creates full coverage and the freedom to play whatever you’d like.

A venue’s requirement to get a compulsory license is one of the most misunderstood aspects of music publishing. The venues themselves especially misunderstand this requirement. I’ve spoken to many business owners who don’t understand why they have to pay for such a license. The phrase “it’s nothing but a shakedown” is used on more than one occasion. But the simple fact is this: if your business is profiting off of someone else’s music – and playing music to enhance your business qualifies – then the songwriters should get a cut of some sort.

There is another argument made by business owners that I find harder to dispute. Nightclub owners often argue that the fees they pay to the PROs aren’t going to the songwriters whose songs they are playing. This statement is often true. Presently, the PROs have no way to track the songs played in their licensed venues. The businesses could submit a list of all the songs played in a day, but no one is going to do that. Instead, the PROs pool the collected fees and distribute the royalty to songwriters they assume are the ones getting played the most. In other words, popular songwriters, for the most popular songs.

I can empathize, as I DJ’ed hundreds of times exclusively at underground clubs and very few of the songwriters I played (if any) ever saw a penny. I’ve heard tales of clubs in some territories tackling the problem by having the DJs write down all the songs from their sets. I guess it’s the thought that counts, but this is obviously an unreliable and haphazard solution.

There’s a change coming, though. Advances in audio recognition are making song tracking in venues possible. Using technology popularized by the likes of Shazam, songs get identified and, in turn, the appropriate songwriters paid. From a story in Complete Music Update:

Collecting societies PPL and PRS For Music have confirmed that they are expanding a pilot project to test the use of music recognition technology in clubs, pubs, bars and hotels to monitor what music is being played in those spaces.

Peter Marks {CEO of UK clubbing chain The Deltic Group} has welcomed the pilot, saying: “Music is the very heartbeat of our business and it’s in our interest to see that talented artists are rewarded for their creations. With online streaming and other digital technology, it’s increasingly difficult for songwriters and musicians to make a living from their creations, so anything we can do to help and attract and support the latest local talent has to be a good thing”.

I believe GEMA in Germany has also been testing this out.

The ramifications are enormous and welcome; accurate tracking in venues (and eventually across other outlets such as radio and sporting events) will create a great benefit for non-mainstream songwriters.

It remains to be seen if US PROs might look to adopt this technology. The fact that there are multiple PROs in this country may prove to be a stumbling block. A device that listens, identifies songs, and sends data to the PROs would have to be installed in every participating venue. It would be a hassle if each PRO had its own device for every business to install. Could they agree on one shared device? Part of me thinks it unlikely as the US PROs are fiercely competitive. That said, the recent news of ASCAP and BMI collaborating on a musical works database gives us a glimmer of hope.

The US is often the country left behind when it comes to advances in rights management. Let’s hope our industry is proactive in embracing this technology solution to a longstanding problem.

If You Are Losing the Game, Best to Change the Rules

· Leave a Comment

Ben Thompson for Stratechery:

The long-rumored competitor to Amazon Echo and Google Home was, fascinatingly, framed as anything but. {Tim} Cook began the unveiling by referencing Apple’s longtime focus on music, and indeed, the first several minutes of the HomePod were entirely about its quality as a speaker. It was, in my estimation, an incredibly smart approach: if you are losing the game, as Siri is to Alexa and Google, best to change the rules, and having heard the HomePod, its sound quality is significantly better than the Amazon Echo (and, one can safely assume, Google Home). Moreover, the ability to link multiple HomePods together is bad news for Sonos in particular (the HomePod sounded significantly better than the Sonos Play 3 as well).

Of course, superior sound quality is what you would expect from a significantly more expensive speaker: the HomePod costs $350, while the Sonos Play 3 is $300, and the Amazon Echo is $150. From Apple’s perspective, though, a high price is a feature, not a bug: remember, the company has a hardware-based business model, which means there needs to be room for a meaningful margin.

The result is a product that, beyond being massively late to market, is inferior to the competition on two of three possible vectors: the HomePod is significantly more expensive than an Echo or Google Home, it has an inferior voice assistant, but it has a better speaker. That is not as bad as it sounds: after all, the iPhone is significantly more expensive than most other smartphones, it has inferior built-in services, but it has a superior user experience otherwise. The difference — and this is why the iPhone is so much more dominant than any other Apple product — is that everyone already needs a phone; the only question is which one. It remains to be seen how many people need a truly impressive speaker.

Coming from a music industry POV, an emphasis on sound quality as a feature – as it applies to music playback – is a great move and may even raise the bar for competitors’ forthcoming hardware. I look forward to personally assessing just how good this HomePod speaker sounds, and find it fascinating that the HomePod has been successfully positioned so that the sound quality is what I’m mostly curious about.

Hitting the Links: Music’s Technological History, Repetitive Pop Lyrics, and Peter Saville

· Leave a Comment

Technology In Music: A Chronological Playlist Through History:

Let’s start from from the beginning, in 1937: a timeless feel – eerie and alienating at times, permeates ‘Oraison’ by French composer Olivier Messiaen. The song was originally written for an ensemble of early electronic musical keyboards called Ondes Martenot. The Ondes Martenot is a very expressive instrument, meeting Messiaen’s avant-garde composition techniques. If you’re expecting beat drops you may want to keep in mind the release date.



Peter Saville On His Album Cover Artwork:

{Saville on New Order’s Technique cover:} It was a garden ornament and we rented it for the shoot. It’s a very bacchanalian image, which fitted the moment just before the last financial crash and the new drug-fuelled hedonism involved in the music scene. It’s also my first ironic work: all the previous sleeves were in some way idealistic and utopian. I’d had this idea that art and design could make the world a better place. That even bus stops could be better.



Are Pop Lyrics Getting More Repetitive?:

In 1977, the great computer scientist Donald Knuth published a paper called The Complexity of Songs, which is basically one long joke about the repetitive lyrics of newfangled music (example quote: “the advent of modern drugs has led to demands for still less memory, and the ultimate improvement of Theorem 1 has consequently just been announced”). I’m going to try to test this hypothesis with data. I’ll be analyzing the repetitiveness of a dataset of 15,000 songs that charted on the Billboard Hot 100 between 1958 and 2017.



Forgotten Tributes: 25 Monumental Relics of Yugoslavia:

If you were to travel today through the area that was once Yugoslavia, you would come across some incredible massive objects that allow an unexpected look directly into the area’s past. Before dissolving into several smaller countries in the 1990s, Yugoslavia became home to a number of large-scale futuristic monuments.



Gary Vaynerchuk on Impact Theory (Podcast):

I think that Steve Jobs came along {and} became an icon, but the sad part of that narrative was he did not treat his employees well. He became an icon and the narrative became he got the most out of people by being a jerk, and that became romanticized. And a lot of people in Silicon Valley today run companies where they’re mean because they think that’s the right thing to do because they put Steve Jobs on a pedestal. I want to become that big {but} what I want to come from that is that kids that aren’t even born today think that they can build a five billion dollar company and {still} be a great guy or a great gal. I want to build the biggest building in town ever by just building the biggest building in town, while I think most people try to tear down everyone else’s building.

Music as Data, and the Wisdom of the Crowd

· Leave a Comment

The Conversation:

The Musical Genome, the algorithm behind Pandora, sifts through 450 pieces of information about the sound of a recording. For example, a song might feature the drums as being one of the loudest components of the sound, compared to other features of the recording. That measurement is a piece of data that can be incorporated into the larger model. Pandora uses these data to help listeners find music that is similar in sound to what they have enjoyed in the past.



This approach upends the 20th-century assumptions of genre. For example, a genre such as classic rock can become monolithic and exclusionary. Subjective decisions about what is and isn’t “rock” have historically been sexist and racist.



With Pandora, the sound of a recording becomes much more influential. Genre is only one of 450 pieces of information that’s being used to classify a song, so if it sounds like 75 percent of rock songs, then it likely counts as rock.



Meanwhile, Shazam began as an idea that turned sound into data. The smartphone app takes an acoustic fingerprint of song’s sound to reveal the artist, song title and album title of the recording. When a user holds his phone toward a speaker playing a recording, he quickly learns what he is hearing.



The listening habits of Shazam’s 120 million active users can be viewed in real time, by geographic location. The music industry now can learn how many people, when they heard a particular song, wanted to know the name of the singer and artist. It gives real-time data that can shape decisions about how – and to whom – songs are marketed, using the preferences of the listeners. Derek Thompson, a journalist who has examined data’s affects on the music industry, has suggested that Shazam has shifted the power of deciding hits from the industry to the wisdom of a crowd.

Discovering Electronic Music: A Short Documentary From 1983

· Leave a Comment

Electronic Beats:

In his documentary “Discovering Electronic Music” director and writer Bernard Wilets explores the basics of early analog synthesizers and the first digital sampling techniques. With its dreamlike and slightly dated approach, it’s a worthwhile watch— and if you’re curious about how future technology was referenced in the past, this short documentary is every paleofuturist’s dream.



Update: Here’s a handy GIF I created from this video.

Crudman: New Musical Instrument Based On A Hacked Walkman

· Leave a Comment

Boing Boing:

The Crudlabs Crudman is a new tape-based instrument consisting of a hacked cassette Walkman controlled by pressing notes on a keyboard. It might remind you of a Mellotron, another tape-based keyboard popular in the 60s and 70s, but Crudlabs points out that the Crudman “is not designed to replace a Mellotron and it does not sound like a Mellotron.”



“The Crudman can provide endless atonal sounscapes but has been designed specifically to be just accurate enough to function as a traditionally melodic musical instrument. You can record anything onto a tape, so if you’re a fan of the sound and idiosyncrasies of tape, the possibilities are pretty much endless.”


WANT. This is pretty great, though having multiple units (at $375 a pop) for polyphony seems to be an expensive requirement for those Boards Of Canada-ish pads. I’m quite curious just how this works so I plan on digging into the documentation found here. And the product’s main site has more video of the Crudman in action. If you can’t afford this lo-fi sweetness, then perhaps you could just pepper your samples with Goldbaby’s Dirt and Layers pack. (My teenage self – who regularly lost sleep over an inability to completely remove tape hiss in his four track demos – is giving me the stink eye.)