I wanted this for the Mac, where there was no low-level way to access the mic and speaker hardware like on the PC, so I was concerned that there wouldn't be a fixed time offset available with Apple APIs.Īt the time, I thought they were just dismissing me as a n00b, but now I understand that they were just terribly overextended. I was thinking that maybe they would let it dynamically adjust with little overhead, rather than needing a fixed delay between the microphone and speakers.
I suggested some stuff with autocorrolations to find the echo delay offset, since I had used them quite a bit as a contractor at hp. I found the Ogg maintainers to be rather hostile to my suggestions around echo cancellation. When I was working on a networked game around 2005, I dabbled with Speex and Vorbis. But in this case, it looks like it has more to do with difficulties around maybe Apple's secrecy and patents around their own echo cancellation implementation. Normally I would attribute this to these types of libraries evolving from the UNIX mindset, which almost always lacks sane defaults. I checked and they do have 2 types of echo cancellation (mixed/mono and multichannel):īut looks like they maybe still have blockers that prevent them from enabling it by default: People sacrifice a good bit of everything else, just to get a better UX, and the more developers realize this, the more potential their projects have to succeed. But if someone wants to make FOSS successful, they better focus on the user experience.
#Mumble echo software
I can see FOSS software outperform Discord, if you pick an aspect, like "low latency audio". With a buttery smooth transition to both.
#Mumble echo download
And you have several ways to "upgrade", you can register your handle, download and use the desktop app. You click on an invite link, it only asks you a handle, and bam, you land on the server, ready to go. The other thing is that joining is like zero effort.
For one, Discord must have a serious backend and constant support to back this all up. Surely it was a bit of time to get the hang of their UX, but I can't seriously see _any_ FOSS software to do audio, video, screen sharing, IM with groups and users, all packed into a nice consistent package. Discord is featureful and each feature is only a couple of clicks away, and every nitty-gritty is nicely abstracted away under the UI. I don't know how the commenters below you miss your point consistently.
0 kommentar(er)
