... the new spec is introducing the Low Complexity Communication Codec—or LC3, for short—which promises better audio quality at lower data rates. Hardware makers will be able to adapt the codec as they need, but LC3 promises to require about half the bandwidth of the audio codecs Bluetooth uses now with improved sound quality.
~ https://gizmodo.com/the-biggest-problems-with-bluetooth-audio-are-about-to-1840461602 (https://gizmodo.com/the-biggest-problems-with-bluetooth-audio-are-about-to-1840461602)
Here is some info about LC3 format
https://www.etsi.org/deliver/etsi_tr/103500_103599/103590/01.01.01_60/tr_103590v010101p.pdf
The biggest problem with Bluetooth Audio is that there is no lossless mode :))
Is there an encoder available to test out the LC3 codec's audio quality?
I wonder why they didn't just use Opus instead of creating a new codec?
I wonder why they didn't just use Opus instead of creating a new codec?
Skimming the etsi.org PDF, LC3 outperformed Opus at 32 kbps in their tests. But they used an old version (1.1.4) with encoding complexity set to 0.
I wonder why they didn't just use Opus instead of creating a new codec?
If you look at the specification, it's basically Opus with a few marginal gadgets thrown in. I'm going to make a not-very-wild guess that you'll need a license for the gadgets, from the people who wrote the standard.
the biggest problem isn't the codec, SBC is not that bad if used properly
IMO the biggest problem with Bluetooth is latency and from what I see in that PDF, this has been at least improved significantly with the LC3 codec, compared to SBC. I doubt you'll be able to play a MIDI instrument with no perceived delay, but it should at least be good enough for most interactive use.
The second biggest problem is the inability to stream to multiple devices and this has also been addressed.
So it looks kinda promising, but let's wait and see how it works in practice.
I wonder why they didn't just use Opus instead of creating a new codec?
Skimming the etsi.org PDF, LC3 outperformed Opus at 32 kbps in their tests. But they used an old version (1.1.4) with encoding complexity set to 0.
My quick skim looked like all of their tests were with speech only, not music. Did I miss it?