[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Linphone-users] call recording experience report
From: |
Greg Troxel |
Subject: |
Re: [Linphone-users] call recording experience report |
Date: |
Sat, 26 Jun 2021 07:48:43 -0400 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/26.3 (berkeley-unix) |
Dennis Filder <d.filder@web.de> writes:
> On Sat, Jun 19, 2021 at 02:49:00PM -0400, Greg Troxel wrote:
>
>> That was an unfamiliar extensions, so:
>>
>> https://en.wikipedia.org/wiki/Matroska
>>
>> and that's philosophically great, but a lot of complexity for what could
>> be just a single audio file. I started to build mkvtoolnix, but it
>> seems to take hours, and audacity worked, so I stopped.
>>
>> So this is only sort of a bug report, but through it might be helpful to
>> people dealing with trying to record.
>
> Matroska was a reasonable choice given that Linphone is also a video
> phone, and needs to be able to record both. At the time of
> implementation it was one of the few unproblematic container formats
> wrt. to intellectual property shenanigans and that can also support
> new codecs easily.
Certainly a fair point.
I guess I would suggest that a recording of an audio call just be the
bare file, but I guess the point is that you can start out audio and
then turn on audio later.
I think really the bug is that there isn't a simple, command-line only
way to pull the streams out as separate files, where "simple" means "not
a lot of dependencies", "can be built on a 10 year old computer in a few
minutes", and "needs at most C++11, nothing newer, no boost, no qt". In
all seriousness, I was amazed at how beastly mkvtoolnix was. I use
pkgsrc, a portable packaging system, and it was included, so it was just
"make pakcage-install" to recurse to building dependencies and then it,
but it was seeming like it was going to take hours, on a computer that
can build firefox in an hour or two. I do not understand why.
signature.asc
Description: PGP signature