Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×

Submission + - Opus 1.5 is Out: The Opus Codec Gets a Serious Machine Learning Upgrade

jmv writes: After more than two years of work, Opus 1.5 is out. It brings many new features that can improve quality and the general audio experience through machine learning, while maintaining fully-compatibility with previous releases. See this release page demonstrating all the new features including:
  • Significant improvement to packet loss robustness using Deep Redundancy (DRED)
  • Improved packet loss concealment through Deep PLC
  • Low-bitrate speech quality enhancement down to 6 kb/s wideband
  • Improved x86 (AVX2) and Arm (Neon) optimizations
  • Support for 4th and 5th order ambisonics

Comment Don't remind people (Score 1) 106

No idea on the men-vs-women thing.

But it seems absolutely crazy for the DRMed media sales industry to remind people that their media could Just Work and be normal, instead of requiring specific proprietary players (a different one for each media source). They shouldn't even mention piracy, because that just plants the seed that people could instead have standard format files, where things are much more convenient than the awkward situation with DRMed media.

If we want people to just accept that things are shitty and must always remain shitty, then it's probably best to not encourage people to think about the topic at all. Shhhh! Don't bring it up, and pretend that the idea of a convenient media library, where users have the choice to use whatever player software that they want on whatever device that they want, simply doesn't exist at all.

Comment Re:Different level of programming (Score 2) 43

Branches are particularly bad on GPUs. In that SIMD model of computation you usually have multiple hardware blocks (called threads but unrelated to CPU threads) implementing the same instruction on different data in parallel. Since some of the threads will evaluate the branch condition to true and others false, and they are all operating in lockstep, that means the hardware needs to execute both branches every time, with some of the threads just sitting doing nothing while the other threads execute the "if", and visa-versa while the others execute the "else".

At least that was the case when I was doing CUDA 5+ years ago. Don't know if things have changed since then.

Comment Cool, I guess (Score 1) 70

This reminds me of how in the 1980s, things like FPUs and MMUs were separate chips. Do you want an 80387 with your 80386? Do you want a 68851 with your 68020? But then the newer CPUs just came with that stuff.

Even if 90% of the machines sold over the next few years never use it (think of how many 80386 chips were running MS-DOS as a "fast 8086" and never went into protected mode), it's nice that on the software side you'll eventually be able to expect it. In 1988 you couldn't assume floating point was fast for everyone, but by 1998 you could.

Comment Re:gross EU government (Score 3, Insightful) 20

Google is free to completely ignore these bullshit requirements and stop doing business in Europe.

For whatever reason, they have chosen to keep transacting with Europeans. Perhaps they chose poorly, and should have instead consulted Slashdot posters about whether or not making tons of money is worth the outrageous indignity.

Comment Re:Ya don't say (Score 1) 40

Doesn't adding the disclaimer truly fix the problem, though? Apparently nontechnical users didn't understand what incognito does, so a sufficiently-well-written disclaimer ought to be able to fully correct the misunderstanding.

On the techie side, we all know that a browser setting isn't going to somehow magically keep other peoples' computers from remembering users' requests, but non-techies didn't understand that magic isn't a thing, so Google's understandably under some pressure to better-document the incognito feature.

Comment All the same problems as DRM (Score 1) 67

Imagine the [unlikely?] case where someone wants to implement FACstamp on their own computer. Can they?

They'd end up facing a similar problem as DRM standards: whoever backs it can't allow any independent implementations, because that would undermine the purpose: preventing people from signing the "wrong" data.

So this FACstamp idea requires proprietary software for every step of the process, with a key obfuscated or hidden inside a TPM chip or something like that. Wanna write something that is interoperable with it? You can't.

Comment Re:Alternative (Score 1) 196

Authors can license textbooks instead of selling them, but do they?

I guess I wouldn't be surprised if kids these days (yes, I'm old) are agreeing to EULAs when they open their textbook apps. But I know for sure that tens of millions of people still alive today, purchased textbooks instead of licensing them. If those textbooks still exist, then the knowledge is attainable without any contracts, so there's no means of discriminating against computers.

Just avoid the weird textbooks (ones that require special software to read) and anyone's LLM can get around the problem you're describing.

Slashdot Top Deals

The Tao is like a glob pattern: used but never used up. It is like the extern void: filled with infinite possibilities.

Working...