Donate Your Noise To Xiph/Mozilla's Deep-Learning Noise Suppression Project (xiph.org) 119
Mozilla-backed researchers are working on a real-time noise suppression algorithm using a neural network -- and they want your noise! Long-time Slashdot reader jmv writes:
The Mozilla Research RRNoise project combines classic signal processing with deep learning, but it's small and fast. No expensive GPUs required -- it runs easily on a Raspberry Pi. The result is easier to tune and sounds better than traditional noise suppression systems (been there!). And you can help!
From the site: Click on this link to let us record one minute of noise from where you are... We're interested in noise from any environment where you might communicate using voice. That can be your office, your car, on the street, or anywhere you might use your phone or computer.
They claim it already sounds better than traditional noise suppression systems, and even though the code isn't optmized yet, "it already runs about 60x faster than real-time on an x86 CPU."
From the site: Click on this link to let us record one minute of noise from where you are... We're interested in noise from any environment where you might communicate using voice. That can be your office, your car, on the street, or anywhere you might use your phone or computer.
They claim it already sounds better than traditional noise suppression systems, and even though the code isn't optmized yet, "it already runs about 60x faster than real-time on an x86 CPU."
Re: (Score:3)
Re: (Score:2)
Re: (Score:1)
Re: (Score:1)
Dave's not here, man...
Re: EditorDave (Score:2)
Re: (Score:2)
It's even worse for those of us who consume their news through the RSS feed. Every time a typo in a title or body gets corrected, that's a new entry in the feed. And they don't just update the article once, oh no no no! Every typo needs to be discovered independently of the others and warrants its own separate update. I see some articles fly by at least 5 times before the so-called "editors" decide it's finally good enough not to bother anymore, or too old for their readership to still give a fuck.
I can und
Re: (Score:2)
Do they not know how GUIDs work? That article shouldn't come through more than once - it should update your original local copy with the updated version.
Re: (Score:2)
And PS, if you use the URL as a GUID, then it had better never ever change - not change every time the headline gets fixed.
All out of you noise (Score:2)
Re: (Score:2)
Punctuation is missing.
It should be "Donate, you noise."
You noise (Score:2)
You noise? Me Tarzan!
Sounds useful if (Score:4, Interesting)
It says it can remove car noises, but can it remove the audience laughter from the Red Green Show? This is a problem someone needs to solve!
Re:Sounds useful if (Score:4, Funny)
Have you tried duct tape?
Re: (Score:2)
Genius!
Remember (Score:5, Funny)
Remember when Mozilla made a web browser? Pepperidge Farm remembers...
Re: (Score:1)
Re: (Score:2)
QuickBASIC! Oh please be written in QuickBASIC. It's due man!
I wrote a company website in Qbasic on NT4 back in 1997. It was later converted to C++/linux then to Perl/linux. 20 years later, the company is still in business and although none of the Qbasic is still there, some characteristics of the existing system can be traced back to how it was originally implemented in Qbasic.
Re: (Score:2)
You...wrote...html...in Qbasic. Right.
Yes. Remember, this was 1997. php was still in beta and only used by a small handful of people. Java v1 had just been released. The most popular language for writing web scripts in 1997 was probably shell scripts followed by C++. Libraries for web scripting didn't really exist yet. Everything was done by echoing html to the screen. Also at that time, compilers weren't free and I happened to have a Qbasic compiler and had grown up coding apple basic so basic was the language I was most comfortable wi
Re: (Score:1)
Yes a stellar job at failing on multiple product fronts and making itself increasingly irrelevant.
Re: (Score:2)
Probably the fastest CPU they could get their hands on, to get the best numbers possible
Re:Remember (Score:5, Informative)
(I'm the author of the article)
You may not be aware, but around 10 years ago, browsers started including audio technology. This now includes WebRTC which lets you do videoconferencing in the browser (without Flash). As surprising as it may sound, some people like doing VoIP/videoconference. And those who use WebRTC tend to prefer when their audio doesn't have too much noise. And that is why RNNoise is useful.
Iffy (Score:4, Interesting)
I tried to donate noise; using a mac under 10.12.6. Mic is working fine. Safari asks if it can use the mic. The record button stays in for 60 seconds. The playback produces nothing.
I have great noise sources, and would not mind contributing.
Re: (Score:2)
Some browsers/OS, already have some noise suppression running. That may be why you're not hearing anything on playback?
Re: (Score:2)
Well, it's not the OS, I can hear the noise fine in headphones through the system.
I suppose it could be Safari, which is a black box.
Re: (Score:2)
I tried to donate noise; using a mac under 10.12.6. Mic is working fine. Safari asks if it can use the mic. The record button stays in for 60 seconds. The playback produces nothing.
I have great noise sources, and would not mind contributing.
Go to System Preferences | Sound. Select the 'Input' Tab, and de-select the button at the bottom, next to where it says, "Use ambient noise reduction."
Otherwise, you get two seconds of noise recorded, before the MacOS noise reduction on the Mic kicks in. It happened on my first recording.
Re: (Score:2)
No, there is no such "Use Ambient Noise Reduction" check. This is under OS X 10.12.6.
Here's the prefs panel [fyngyrz.com]
Re: (Score:2)
No, there is no such "Use Ambient Noise Reduction" check. This is under OS X 10.12.6.
Here's the prefs panel [fyngyrz.com]
MacOS 10.10 has the checkbox.
My 10.12 machine was stolen, so I couldn't check it. But I'll bet it's still in there, even if only as a 'hidden' preference, like so many other OS X tweaks. Headphone-less FaceTime/video chat/ voice chat works fine -- with no feedback or hums -- and iPhones on speaker-phone exclude ambient noises very effectively. Either is good enough to serve as an 8-person video conference with no yelling required. Plugging in a quality microphone, say a $150 pro-(con)sumer mic and the
Re: (Score:2)
There is no such "Use Ambient Noise Reduction" check. OS X 10.12.6.
Here's the prefs panel [fyngyrz.com]
Re: (Score:1, Flamebait)
Nobody wants video conferencing in their browser and no one fucking uses it either.
Re: (Score:2)
The dwindling user numbers speak for themselves, down to 6% now.
Re: (Score:2)
Quite the opposite. No one wants video conferencing in proprietary apps using a proprietary protocols complete with the wonders of all modern technologies : data mining and ads.
No one uses WebRTC because it is in its infancy. In the mean time many of us are cheering this on from the sidelines.
Re: (Score:3)
No one uses WebRTC because it is in its infancy.
Every time I've tried to use it from behind a NAT I've had issues. It turned out to be the same issues that I had with SIP - not being able to get to the client due to a multi-layered NAT (two NAT layers or more). Most cell providers do double-NAT because for normal browsing and downloading it works just fine.
If you're doing SIP (or something like it, like WebRTC) you need an external proxy (STUN, TURN or ICE) server which all participants talk to, but which still won't work to get through the double-NAT.
W
Do the needful (Score:2)
Re:Data Set Publicly Available? (Score:5, Informative)
The entire project is on github. [github.com]
I found this by going to the link in TFS.
Re: (Score:3)
That link has only the source code. It does not include the training data set.
The submission link requires CC-0 attribution, which makes me hopeful that they plan to release the data freely. But I hunted all over the site and couldn't find either a link to the data or any comment about their plans for it going forward.
Re:Data Set Publicly Available? (Score:5, Informative)
Now, training is a little trickier because I cannot share the data.
I cannot share the current data I'm using because it's copyrighted. Hence asking for people for help getting data that I can redistribute.
So weâ(TM)re supposed to just give jmv a bunch of data with no way to know how he is using it?
Yes, because I have such a track record [speex.org] for keeping [opus-codec.org] things private [github.com].
Unnecessary sarcasm sunk to his level (Score:1)
Had you just posted the links, you'd have secured the moral high ground and made GP look like a fool and a jerk.
By packaging your reply in sarcasm, you gave up the high ground.
Re: (Score:2)
He did no such thing, and judging someone based on their level of sarcasm when dealing with a troll rather than their actions just shows everyone what you consider more important.
THAT's a good way of giving up any moral standing.
Re:Data Set Publicly Available? (Score:4, Informative)
It's the first time we try this. We'll look at the quality of the data we get (yes, noise quality!) and if it's sufficiently good/useful, then we'll also make it available. It might take some time to sort out the useful samples from the ones that aren't since some already have noise suppression applied by the OS or browser.
curl slashdot.org noise.txt (Score:1)
curl slashdot.org > noise.txt
Shouldn't take more than a day or so to have all the info they will ever need.
x86 CPU? (Score:2)
"it already runs about 60x faster than real-time on an x86 CPU."
I'll get my 8086 based XT out of storage, should be perfect.
Re: (Score:1)
Maybe it has excellent scaling?
Re: (Score:1)
On an x86 CPU, not all x86 CPUs. There exists an x86 CPU on which it runs 60x faster than real-time.
Re: (Score:2)
Since when is ‘an’ equivalent to ‘all?’
Re: (Score:2)
Well that's more or less my point - though it was a vague attempt at humour also - the original is a statement which gives us no real information. If the x86 was in fact an 8086 we'd be thinking that on modern hardware it'd be virtually unnoticeable as a workload, if on the other hand it was a i9 using all threads, then perhaps most peoples experience wouldn't be so great.
What is your favorite noise? (Score:3)
I suggest piping in a few tracks by SPK, in particular "Emanation Machine R.Gie 1916", the first track from their 1981 release "Information Overload Unit".
https://www.youtube.com/watch?v=G9b89PFYZ5g
When my wife first heard it, she said it was like having your head stuck inside a running vacuum cleaner. Follow it up with some Throbbing Gristle, perhaps.
Re: (Score:2)
Nice opportunity passed by this weekend - the Knob Creek Machine Gun shoot. Plenty of noise.
Supercomputing Mass Stupidity (Score:1)
You would need a supercomputing cluster to filter out what I would consider "noise" these days, which would include 95% of the pointless shit that makes up social media.
Guess I'm too old to help.
(CAPTCHA: supports)
Grab a mic and pull my finger (Score:2)
Re: (Score:1)
Did you go off your meds again, grandpa?
Re: (Score:1)
Cool story.
The Best Noise Suppression... (Score:2)
Is to have enough signal in the first place to boost the SNR to the point where the noise becomes irrelevant. Good quality worn microphones close to the mouth, dual mic setups for simple background suppression. Sound engineers have known this forever.
Re: (Score:1)
Yeah because everyone is going to buy two mics to make VoIP calls... *rolls eyes*
Re: (Score:2)
That's a great idea - I'll get a lectern and some mics setup in the corner of my home office. Does Elbonia or Durkadurkastahn have a flag? I could put one of those behind me. My morning standups are never going to be the same again, dotards!
Harleys (Score:2)
Can it remove the noise from a Harley-Davidson running straight pipes?
Millions of Husbands... (Score:3, Funny)
Nice try, NSA! (Score:2)
You need to find more clever methods to grab the noise of my PC fan [wired.com], however.
I recorded my bedroom (Score:2)