This is a derivative of my own idea from yesterday [1], inspired by the onscreen note at [2].

It's a video camera, or postprocessing software, that recognizes what song is being played in the surroundings and applies an inverted signal of that same song, derived from a streaming service or offline library, to cancel it out. Then you can more clearly hear other sounds like people speaking. Also, if you post the video on YouTube, you don't have to mute the audio or risk getting it copyright-struck. (Well, you risk that even if you post a video that's entirely black and silent, but it should reduce the risk somewhat.)

N/A [2019-04-01]
-- notexactly, Apr 01 2019

[1] Song-recognizing-and-canceling headphones
My idea from yesterday that this is derived from [notexactly, Apr 01 2019]

[2] William Osman: I'm Still Alive (@9:56) https://youtu.be/V0yGbpUeW7s?t=596
Probably not the best first video from him to watch [notexactly, Apr 01 2019]

Oops—I deleted the previous annotation here, intending to reply to it! If you remember what you said, please post it again.

(If there was an option to do so, I would disable my ability to delete others' annotations on my ideas, to prevent such mistakes.)

All I managed to read before I hit the wrong button was a question whether this used or was derived from a fiber-optic microphone, and a question as to how it recognizes songs. For the first of those, no—it uses any kind of microphone a video camera commonly uses. For the second, it uses an online service such as SoundHound or Shazam (especially easy if the camera is a smartphone, or if the recognition and canceling is done in post) or a local database of music that gets searched through locally (only practical in post, on a pro editing workstation/server).
-- notexactly, Apr 01 2019


I bet it was deliberate
-- pocmloc, Apr 02 2019


I'd be happy if this was a default setting for the world.
-- UnaBubba, Apr 03 2019



random, halfbakery