I think it is very interesting that so many of the early applications of computer technology have to do with audio. John Bardeen's music box, the first commercial application of the transistor in hearing aids, the HP garage in Palo Alto was originally building audio oscillators, the iPhone evolved from the iPod, the internet was built on copper made to carry analog telephone calls, Bell Labs (ping!), the list goes on.

A friend of mine has the hypothesis that maybe human beings end up figuring out how to do kHz stuff before they go on to do MHz/GHz stuff. Not a perfect explanation but kind of attractive...

You can transform every audio signal into an electronic signal relatively easy - for graphics there is so much more complexity involved just in making them visible.

A speaker that translates electronic signal into sound waves is a super simple contraption at its core.

Edit:/ and audio is striking - it has a profound effect on every human (except deaf of course). If I wanted to demonstrate the power of electronics/computers I would choose audio as well.

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”

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”.

(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 ;)

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).

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?

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

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

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.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?

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.

  [1] https://github.com/cgzirim/not-shazam?tab=readme-ov-file#resources--card_file_box

For a while, it seems like Soundhound was about to shut down. It wouldn't match anything released in the last 12-18 months, but that seems to be fixed.

The conclusion appears to be that BeatFind and Shazam know the most songs, but are also somewhat complementary and all of the services had at least one song they uniquely recognised.

---

    Fun facts:
     * Night Driver (W) said "1 Shazams". I think I was the first person to ever Shazam that. Some of the most obscure things had hundreds, often thousands of shazams!
     * You know where they are taking the hobbits but none of the services do!

    ========

    - ABC = found the song
    - # = number of attempts
    - f = exceptionally fast matching (when it did match, might not be first attempt)
    - ~ = knew one of the songs

    BeatFind:     2B  C     1E    2G 2H  1I  2Jf 2Lf 4M 2Nf 2Of 1Pf 1Rf ~S 2T  1Uf 1Vf 1W 1Xf Y Z
    SoundHound: A 1B        1E 2F 1G 2H  2If     1L         2O  3P  2R     1Tf 2U
    Shazam:       1Bf    1D 1E 1F 1G 1Hf 1I  2J  1L  1M 4N  1O  1Pf 1Rf    1Tf 1Uf 1V  2W 1X  Y Z
    MusixMatch: crashes on startup, presumably it realizes it won't be able to show me ads

    missingno
    Shazam:     A C K Q S
    SoundHound: C D F J K M/N Q S V W X Y Z
    BeatFind:   A D F K Q

    non-universal finds (repeated letter = unique = counts double; slash means same artist so should be counted as one)
    Shazam:     DD F J M/N V X Y Z
    SoundHound: AA F
    BeatFind:   CC J M/N S V X Y Z

    A: Levan Polkka Epic Orchestral Cover version
    B: Pokémon red/blue soundtrack
    C: Mayhem (various songs, it seems either they have all or they have none)
    D: Art Now ft. Snowflake
    E: Hero's Choice
    F: Three Days Grace - Scared
    G: Syrian - Supernova
    H: The Explosion - Here I Am
    I: The Von Bondies - C'mon C'mon
    J: Frank Klepacki - Scouting (C&C TibSun)
    K: Conspiracy - Chaos Theory (demoscene)
    L: Cheshyre - Madness6 (remix) (Newgrounds ID 77998)
    M: Dimrain47 - Twilight Techno
    N: Dimrain47 - Cloud Control
    O: DragonForce - My Spirit Will Go On
    P: Yuki Kajiura - The First Town (SAO)
    Q: THEY'RE TAKING THE HOBBITS TO ISENGARD! THE HOBBITS- THE HOBBITS- TO ISENGARD! TO ISENGARD!
    R: Faithless - Insomnia
    S: Age of Empires 1 soundtrack
    T: Moulin Rouge - El Tango De Roxanne
    U: Van Canto - Master of Puppets
    V: Slack Bird - Jouni
    W: Floppytronic - Night Driver
    X: EgoSalad / Kitboga - Breathe in
    Y: Floppy Drive music: top 4 hits on yt: sweet dreams, imperial march, ghostbusters, beat it. Only ghostbusters was known to any
    Z: Obsidian Shell - Orphanage

Note that what I did not test/review introducing noise (like people talking through it) or filtering (like when you hear the music through a wall)

- The instructions seem not to be the best to get it up and running (e.g. "cd not-shazam" and just a few lines later "cd not-shazam/client")

- MongoDB is needed but information on how to hook it up/use it are absent (I would make the DB swapable and provide something less intrusive like sqlite)

- If replacing MongoDB is not possible, I would provide a dockerfile and a docker compose to allow easy startup and testing.

- The client npm install has 8 critical vulnerabilities, these might not actually matter but it makes me hesitant to continue testing

- You might not care about the patent or the copyright, but I would still change the name at the very least. Github itself is located in the US and will remove the project if they receives a DMCA.

- Last, this might not be as important, I would add a way to add songs from wav files. Not everything I'd want to test this with is on spotify or youtube.

I'm not saying this to discourage you or anything, I just think the project needs that little extra bit of polish. Minor things will cause people to discredit or ignore a project. If I get around to it I might make a PR for the project. I want to experiment with audio matching outside of the music space, and your project seems like it'll be the easiest to modify.

Edit: Formatting

I agree that the project could definitely use some polishing. I'll prioritize improving the setup instructions and look into adding a file-based DB for flexibility, as well as resolving the npm vulnerabilities. Adding support for directly fingerprinting wav files is a great idea and something I'll prioritize, too.

Regarding the project name, I understand the potential legal implications and will definitely change it. I'd appreciate any suggestions you might have.

I'm excited about the possibility of your contributions. Please, feel free to open a PR whenever you're ready.

Thanks again for your feedback!

Or is the algorithm more feature hash than a clusterable feature vector?

Using exemplar fingerprints, a representative sample of an artist's music, is a good approach, but success would require detailed fingerprints, a varied dataset, and a well-chosen algorithm.

For artists who change styles, time-series analysis can capture their evolving sound.

The solution will likely need machine learning.

The current solution doesn't use feature hashing or clusterable feature vectors. Instead, it relies on audio fingerprinting, which breaks down short audio samples into unique patterns or "fingerprints" for quick comparison with a large database of known songs.