27 thoughts on “Sound Recognition – Computerphile”

  1. He touched on a thing that drives me crazy. I've always found foley design fascinating, ever since I found out that the sound you hear when a punch lands in TVs or movies or videogames is actually the sound of celery snapping. What drives me crazy is that when, say, a videogame includes a couple animated characters trading blows with punch sound effects, people refer to that as "realistic" violence. They can get quite worked up about it, but all they do is illustrate either how tremendously ignorant they are about reality or else how effective foley designers (and other creative types) in the other media they've consumed have been in feeding them fiction that they now accept as more 'realistic' than reality is. We're at a point now where most adults literally can not tell the difference between fiction and reality in a very real way. Every time you hear people nattering on about needing to censor games, movies, TV, etc or keep them away from children, this is what you're hearing. Grown adults flaunting that they either don't know what reality is like or that they're so completely taken in by some utterly un-real visualizations and foley effects that they've lost their grip on what is real and what isn't.

    Also, speech recognition is only done in the 'cloud' because Amazon, Google, Apple, etc mine the data for business purposes. A trained recognition neural network does not need significant computational power to actually run. The learning takes big power, but once the system is trained it can be run extremely easily on cheap commodity hardware. The corporations, however, primarily make their money by pimping out the data about their customers, so they want to know what you are talking about and thinking about as often as possible so that they can sell the information to advertisers.

  2. 15mins of explaining how you need to cast into a machine learning problem instead of doing anything smart yourself… Not very engaging in my opinion.

  3. The Ad before this video was about IBM's WATSON, saying the elevator will break in the next 2 days.
    That is pretty creepy given that is way too similar HAL 9000 in 2001: A Space Odyssey.
    We are teaching a computers to recognize sounds…
    Pretty soon, SkyNet will become self aware…

  4. i am a hobbyist music maker, and I have kind of created my own dictionary around describing sounds. for example, "smooth, deep, dark blue, hollow, rubbery", or "smooth, sharp, narrow, high, reddish/brown", and such đŸ˜€
    they would be highly insufficient for sound recognition purposes, but I would not be surprised if they were at least partially overlapping with the features they used. as far as i know, many people who make music, at least those who make it on computer so they have to think about these things consciously, have similar kinds of more or less developed dictionary to describe sound, melody, and rythm features.

  5. hey next time can you guys please talk about the gunshot location systems in cities with high violent crime rates it's a pretty interesting concept

  6. I was disappointed by this video. This is an interesting topic, for which there were really two viable approaches for an entertaining 15-minute video: A high-level overview of the technology this guy developed, with diagrams of the parts of a specific use case or device, "how it works" style; or the first video in a series like you've done with Dr. Pound that cover prerequisite fundamental concepts first, and build up to a more in-depth examination of an end-to-end example. This video isn't either; it's 15 minutes of a guy speaking at such a high-level of abstraction that nothing is actually explained, because he ostensibly thinks that computerphile viewers aren't sophisticated enough to understand his work, as is made evident by his condescending remark at 8:33 – 8:40. How about an example of a feature that IS useful, and some examples of what exactly the metrics are, and what sounds it can be used to differentiate?

  7. I was using Shazam on a episode on 'salvage hunters' where a guy in a church was playing live on a organ. It found the number as 'Toccata symphony no. 5'. Not shure what going on, maybe other has searched and the pattern has been added to the database.

  8. Everyone who played DOOM 1 and 2 in the 90s knows how many sound effects of these games they adopted into various movies, it's kind of weird. Or does anyone remember the typical sound that computers made in the 80s and 90s, something like a squeaking clicking noise that shall remind the audience of a harddrive without actually sounding like a hard drive…

  9. I would think designing a security system based on audio would be challenging as an attacker could simply jam the classifier with extremely loud, rather short duration wideband noise to temporarily raise the noise floor. I can think of a few mitigations, but I'd imagine it'd be pretty hard to design a system with audio as the only sensor.

  10. I have a patent for sound recognition that matches a prerecorded audio template of either vocalized sounds or speech with a live
    audio template. If the live audio sounds like the prerecorded audio template then there is notification. Speech recognition, such as "Alexa" or "Siri" are manufactured by established companies such as Google or Apple, but the sound recognition technology is still in it's infancy and there aren't many sound recognition manufactures. I'm looking for a manufacturing company that will purchase part or all of my patent. If anyone has any suggestions as to how to find such manufacturing companies with the capability to manufacture the device as a Bluetooth device, device as a cellular a host, GPS capability, and sound recognition manufacturing capability. Please let me know if there are any manufacturing companies that might be interested. Thank

  11. Hey! I can't program at all (yet! ;)). I need smthing that recognises a very specific sound real time. I would be very great full for anyone here who could send me a link, a book or anything helping me out to separate timewasting (even tho I liked this video)

Leave a Reply

Your email address will not be published. Required fields are marked *