Hacker News Reader

Comment by halfmatthalfcat | Hacker News Reader

FYI - If this is a true reproduction of Shazam, it’s under patent by Apple through at least March 2025[1].

[1] https://patents.google.com/patent/US7627477

“An Industrial-Strength Audio Search Algorithm”, the paper where Shazam describes their algorithm (https://www.ee.columbia.edu/~dpwe/papers/Wang03-shazam.pdf) doesn’t have a clear publication date, but https://www.researchgate.net/publication/220723446_An_Indust... indicates it is from 2003.

That patent was filed in the US on 2004-10-21.

IANAL, but to me, that’s a point against that patent in the USA.

How is it a point against the patent if the paper was written by someone under the employment of Shazam? Isn't the point of the patent to award the innovator with the right to profit from the innovation?

You only get a one year grace period after first public discloser to file for a patent in the US. So if the dates in this scenario are: - paper in 2003 - patent 2004-10-21

Only if the paper was released between 2003-10-22 and 2003-12-31 would it meet the one year grace period requirement.

Looks like it’s not relevant in this case since they got a provisional patent in 2002, but that’s likely what the above was referring to as “a point against that patent”.

Thank you! That's elucidating

That it was written by someone under the employment of Shazam makes it likely that it describes their algorithm, but for patent protection, what matters is that you can’t apply for a patent for an invention that has been published.

https://www.science.org/content/article/patent-first-publish...: “According to U.S. law, a patent cannot be obtained if an invention was previously known or used by other people in the U.S., or was already patented or published anywhere in the world. Furthermore, publicly using or selling an invention more than 1 year prior to filing a patent application completely bars you from ever winning a patent on that invention.

[…]

In Europe, for instance, there is no 1-year grace period--the chances of winning patent protection is lost the instant an invention becomes public”

> [T]he point of the patent [is to] award the innovator with the right to profit from the innovation

... in return for disclosing said innovation for use by others (in the patent) instead of keeping it secret. If the disclosure has already happened without the prospect of royalties (such as in a journal article), then the deal is off. But indeed various places have grace periods for that.

Actually the provisional patent # 60/376,055 was filed on April 25 2002.

So if the paper was indeed published in Oct 2003 then all is well.

As long as you have submitted the patent before publication you are good to go.

You should note that the USA even today still has a one year grace period between when you first published an idea and when you can valodly file for a patent for that idea. So if they published it in June 2003 but applied for the patent in April 2004, then the patent is well within the grace period and the paper doesn't constitute prior art.

I remember a popular HN post from 10 years ago, that was pulled or the source was pulled because Shazam legally threatened the disclosure of the algorithm. I think it's actually the Google drive file pdf capture from OP's article.

Here is a link to that HN thread from 2013:

https://news.ycombinator.com/item?id=5723863

Related. Others?

How Shazam Works (2003) [pdf] - https://news.ycombinator.com/item?id=40029036 - April 2024 (29 comments)

How does Shazam work? (2022) - https://news.ycombinator.com/item?id=38531428 - Dec 2023 (154 comments)

An Industrial-Strength Audio Search Algorithm (2003) [pdf] - https://news.ycombinator.com/item?id=33299853 - Oct 2022 (1 comment)

Creating Shazam in Java (2010) - https://news.ycombinator.com/item?id=32530056 - Aug 2022 (36 comments)

Shazam turns 20 - https://news.ycombinator.com/item?id=32520593 - Aug 2022 (227 comments)

How Shazam Works (2015) - https://news.ycombinator.com/item?id=23806142 - July 2020 (7 comments)

Designing an audio adblocker - https://news.ycombinator.com/item?id=18855029 - Jan 2019 (186 comments)

Show HN: A radio/podcast adblocker featuring ML and Shazam-like fingerprinting - https://news.ycombinator.com/item?id=18459058 - Nov 2018 (2 comments)

Apple has completed its acquisition of Shazam - https://news.ycombinator.com/item?id=18066724 - Sept 2018 (316 comments)

Apple Buys Shazam to Boost Apple Music - https://news.ycombinator.com/item?id=15899065 - Dec 2017 (156 comments)

Apple is close to acquiring Shazam, sources say - https://news.ycombinator.com/item?id=15881896 - Dec 2017 (292 comments)

Show HN: Shazam-like acoustic fingerprinting of continuous audio streams - https://news.ycombinator.com/item?id=15809291 - Nov 2017 (76 comments)

How Shazam Works (2015) - https://news.ycombinator.com/item?id=15350729 - Sept 2017 (13 comments)

Tell HN: Shazam picks up song from my kitchen light - https://news.ycombinator.com/item?id=11593305 - April 2016 (2 comments)

How Shazam works - https://news.ycombinator.com/item?id=9870408 - July 2015 (48 comments)

Patent infringement claim re: “Creating Shazam in Java” blogpost (2010) - https://news.ycombinator.com/item?id=9594480 - May 2015 (18 comments)

The Shazam Effect (2014) - https://news.ycombinator.com/item?id=9593429 - May 2015 (37 comments)

The Shazam Effect - https://news.ycombinator.com/item?id=8634357 - Nov 2014 (34 comments)

Ask HN: Is there an audio search technology that finds exact and similar audio? - https://news.ycombinator.com/item?id=8420141 - Oct 2014 (3 comments)

Source code example of the Shazam algorithm - https://news.ycombinator.com/item?id=5724442 - May 2013 (16 comments)

Creating Shazam in Java - https://news.ycombinator.com/item?id=5723863 - May 2013 (43 comments)

An Industrial-Strength Audio Search Algorithm (Shazam) - https://news.ycombinator.com/item?id=2621103 - June 2011 (4 comments)

Shazam's Search for Songs Creates New Music Jobs - https://news.ycombinator.com/item?id=2215295 - Feb 2011 (1 comment)

How does the music-identifying app Shazam work its magic? - https://news.ycombinator.com/item?id=2214992 - Feb 2011 (2 comments)

How Shazam Works To Identify (Nearly) Every Song You Throw At It - https://news.ycombinator.com/item?id=1727891 - Sept 2010 (1 comment)

Implementing Shazam with Java in a weekend - https://news.ycombinator.com/item?id=1702975 - Sept 2010 (23 comments)

Shazam: not magic after all - https://news.ycombinator.com/item?id=909263 - Oct 2009 (28 comments)

How does the music-identifying app Shazam work its magic? - https://news.ycombinator.com/item?id=893353 - Oct 2009 (16 comments)

Shazam Has 50 Million Users and Secures Investment From KPCB - https://news.ycombinator.com/item?id=882537 - Oct 2009 (13 comments)

Thanks! Macroexpanded:

Creating Shazam in Java (2010) - https://news.ycombinator.com/item?id=32530056 - Aug 2022 (36 comments)

Patent infringement claim re: “Creating Shazam in Java” blogpost (2010) - https://news.ycombinator.com/item?id=9594480 - May 2015 (18 comments)

Creating Shazam in Java - https://news.ycombinator.com/item?id=5723863 - May 2013 (43 comments)

Implementing Shazam with Java in a weekend - https://news.ycombinator.com/item?id=1702975 - Sept 2010 (23 comments)

Longer list of relateds here: https://news.ycombinator.com/item?id=41132462.

It was a Dutch guy. I have seen his talk on this something with fast Fourier transform and getting sued. Was fun

It's:

(1) deriving a simple fingerprint from the FFT of the audio signal

(2) simple indexing

(3) simple similarity search

You need the signatures of all music on earth for this to work though ;)

>all the music on earth… I would presume that adding a searchable fingerprint to every torrent link would do it

> You need the signatures of all music on earth for this to work though ;)

Not that difficult with a Spotify subscription. It will take some time, though.

Would-be Neil Young fans will be left wanting if Spotify is the only reference.

He's back... are you talking about his archive? https://open.spotify.com/artist/6v8FB84lnmJs434UJf2Mrm

Oh, I didn’t realize he came back. I googled and only found reports of him leaving (I don’t use Spotify).

So not enforceable anywhere else than in the U.S.

There are other patents for their implementation, and all are filed in multiple countries. Look at the blue sidebar under "Worldwide applications".

The linked one is just for the US version of a singular patent. It had applications in 14 other countries and WIPO, 6 of which are still active (plus US).

Any seller of software would be liable for selling software to US customers without a patent license.

I'm curious about the legal consequences of freely distributed software (e.g. open source). I wonder if the author/provider could be held liable if they: - knowingly (passively) or actively market to US customers (e.g. provide support) - are aware that US users are using it, and take no actions to prevent its distribution etc.

Can someone share their knowledge on this?

If European software incidentally infringes a US patent, and it is distributed freely, is the provider then liable? E.g. is Github basically liable for restricting US users from access to (distributing) patent infringing software?

As I understand, you can violate a patent just by importing something infringing. Selling it isn't necessarily a requirement for infringement.

https://blog.shp.law/index.php/2021/03/28/open-source-softwa...

I'm sure there's tons of patent violations in FOSS, but patents are usually just used to go after companies with big pockets.

Due to software patents (generally) not being a thing outside the US, or because it was demonstrated before patent application?

So you are saying to clone the repo now

Figure 1 looks interesting since it has both a time and frequency axis, when usually signals have either a time __or__ frequency axis.

Now I'm curious how the Fourier (?) transform of a signal at a __single__ given timepoint is even defined ...

The concepts you are looking for are the short-time Fourier transform and spectrogram:

https://en.wikipedia.org/wiki/Short-time_Fourier_transform

https://en.wikipedia.org/wiki/Spectrogram

Jamessb already linked to the right terms. One thing to add is that there is always a tradeoff between time and frequency resolution on short time Fourier transforms. You just can’t have both. It’s always a somewhat unsatisfying tradeoff that still works well in practice.

I found this out when I was trying to turn an ordinary 5 dollar thrifted musical keyboard into a midi controller by plugging it into my PC, putting it on "sine wave" and using a Goertzel detector

The latency for detecting audio-frequency waves is quite bad

This also stymied my desire to put digital audio onto a vinyl record :( literally not enough bandwidth

By definition there is vinyl quality worth of bandwidth on the record.

True, honestly the real limiting factor was my signals and encoding lack of knowledge

Soon!

I was under the impression that patents aren’t enforceable on open source projects. I assume I am wrong?

Yes. Just because something is open source doesn't mean it's unbound by IP laws.

It's not whether "IP" laws apply; it's whether source code itself is in scope for patents. Source code is a description of an algorithm, which in principle is what the software patent is supposed to be providing anyway. Patents shouldn't be relevant here for the same reason they aren't relevant to what you write in e.g. a textbook on signal processing where you might find an exact description of how Shazam works.

Compiling and running that source code on your computer/as part of a wider system may violate a patent, but my impression was that patents are not relevant to the actual code. Are there test cases in the US around pure source distribution of a patented algorithm? Particularly post-Alice?

I wonder if the massive amount of open source software can now be used as elaborate 'prior art' in a way that basically invalidates any software patent that is awarded after the source had been made available?

I.e. if any algorithm was already implemented, in some variation, then the patent is not valid?

For example, for the infamous Amazon 'one click purchase', if a similar pattern was used, maybe a 'one click start vacuum cleaner robot', would it that patent then be invalid?

Depends how much money you have.

For example, Tesla just patented the Robotaxi one year ago, despite having open-source solutions like Apollo Self-Driving platform.

Patents are really an obsolete system that favors the super-rich and the lawyers.

I'd be surprised if they patented the general concept of a robotaxi. Almost certainly, it'd be one specific type they they could say they invented.

Patents are enforceable in the US for even personal use without distribution.

I dunno how software patents work but I was under the impression that unless you basically copy paste their code, the courts wouldn't consider it patent infringement as you can't patent the function, but rather the specific thing itself which for software is the exact code itself. But if I'm not understanding something please correct me.

This cannot be true because if you copied a physical design and made a few irrelevant modifications then it's still infringement.

You’re thinking of copyright, which covers a specific creative expression. Patents are more general on how something is done and would cover different code that works the same way.

Ah, you're right, thanks

Overly simplistic: patents is the process not strictly design.

Edit: just to be clear there are design patents too, but I don’t think they’d be granted for software.

Software patents are magic, you just start your process with 'on a computer do X' and because computers are a piece of hardware you can patent anything you like

I'm still upset that my "you can't patent software because of the Curry-Howard isomorphism" legal argument never took off. (Basically, software is equivalent to math, and you can't patent math. Therefore, you can't patent software. QED.)

Reply on news.ycombinator.com