SoylentNews is people
SoylentNews
In this article the authors introduce . . .
PixelPlayer, a system that, by watching large amounts of unlabeled videos, learns to locate image regions which produce sounds and separate the input sounds into a set of components that represents the sound from each pixel. Our approach capitalizes on the natural synchronization of the visual and audio modalities to learn models that jointly parse sounds and images, without requiring additional manual supervision.
The system is trained with a large number of videos containing people playing instruments in different combinations, including solos and duets. No supervision is provided on what instruments are present on each video, where they are located, or how they sound. During test time, the input to the system is a video showing people playing different instruments, and the mono auditory input. Our system performs audio-visual source separation and localization, splitting the input sound signal into N sound channels, each one corresponding to a different instrument category. In addition, the system can localize the sounds and assign a different audio wave to each pixel in the input video.
A video is included along with an explanation of several interesting demos, such as pointing at any pixel to hear the sound from that pixel. Or remixing the volume levels of different musical instruments in the video.
The paper is included along with the data set. It says the code is coming soon.
(Score: 3, Interesting) by FatPhil on Tuesday September 25 2018, @08:44PM (1 child)
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
(Score: 2) by BsAtHome on Tuesday September 25 2018, @09:07PM
That would be technically sound. However, the currently heard hype is neural networks. No money for simple engineering of waves. This is the buzz of played words that has to be split into a network of big computers using more power achieving less. That is where we hear the bells ring.
Now, picture that!
(Score: 4, Informative) by ledow on Tuesday September 25 2018, @09:20PM (1 child)
If could have implications for post-editing, for removing the background from live performances, to even synchronising audio and video visually.
If it wasn't actually quite bad at doing what it does.
(Score: 2) by suburbanitemediocrity on Tuesday September 25 2018, @10:47PM
I think this has been done before...melodyne's "direct note acess" claims to pull individual notes out of a chord and chang it anyway you'd like. It has been around for a decade or so
https://www.youtube.com/watch?v=jFCjv4_jqAY [youtube.com]
(Score: 0, Informative) by Anonymous Coward on Tuesday September 25 2018, @10:37PM (1 child)
AI is really good at that, I hear. Maybe they can tell it to listen for pixels that say "Allah Akbar Jihad Jihad".
(Score: 2) by DannyB on Wednesday September 26 2018, @01:59PM
For your purposes, you could train an AI to recognize terrorists by their skin color.
Infinity is clearly an even number since the next higher number is odd.
(Score: 2) by crafoo on Tuesday September 25 2018, @10:45PM
So is this just bandpass filters and basic image recognition? huh
(Score: 0) by Anonymous Coward on Tuesday September 25 2018, @11:03PM
So can it generate the video that would produce a sound track? Ala winamp visualizations?
Or maybe generate the audio that corresponds to a video?
(Score: 0) by Anonymous Coward on Tuesday September 25 2018, @11:06PM
Dave, although you took very thorough precautions in the pod against my hearing you, I could see your lips move. [youtube.com]
(Score: 2) by takyon on Tuesday September 25 2018, @11:20PM
https://news.mit.edu/2014/algorithm-recovers-speech-from-vibrations-0804 [mit.edu]
https://arstechnica.com/science/2014/08/researchers-reconstruct-human-speech-by-recording-a-potato-chip-bag/ [arstechnica.com]
https://www.npr.org/2016/03/18/470514319/how-can-hidden-sounds-be-captured-by-everyday-objects [npr.org]
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 4, Insightful) by fishybell on Tuesday September 25 2018, @11:53PM (2 children)
I feel like what was missing (didn't RTFA, but did WTFV) is picking out the sound of a person talking and isolate that. Even if all it did was drown out background noise, then we'd have a winner. Adding in existing filtering technologies, and we're talking not just about a potentially profitable product, but also a scary big brother tech that could isolate sounds in a crowd.
(Score: 2) by mhajicek on Wednesday September 26 2018, @12:31AM (1 child)
If I could take a YouTube video and remove the annoying music playing too loudly over the person talking that would be great. Or take a TV show or movie and cut the volume of the music and sound effects, and boost the dialogue so I can actually understand what's being said.
The spacelike surfaces of time foliations can have a cusp at the surface of discontinuity. - P. Hajicek
(Score: 1, Interesting) by Anonymous Coward on Wednesday September 26 2018, @03:33PM
Sometimes also the opposite applies: Music played in the background of some talking, and you like the music, and would like to remove the talking without affecting the quality of the music, so you can enjoy the music in isolation.