@DSPGURU Azid -o switch

[raistlin2k]

Before there was your Besweet I used Azid and oggenc in 2 steps.

There I used following azid-switches on the audiostream containing the directors commentary:
azid -d 1/0 -o C %input% %output%

As you can see, I'm used to downmix it to mono, since it's only speech, stereo is not needed, and mono is twice as fast on transcoding.

Unfortunately, Besweet only has the -d switch, but no -o switch, so it doesn't work.

Could you add this azid-switch??
Would be really great!!

Thanks
Raist

[DSPguru]

i'll consider it.

btw, if you're into trying new stuff, you can try to encode speech with Speex (http://speex.sourceforge.net).
they're now working on packing their bitstream into Ogg packets.

[pacohaas]

Originally posted by raistlin2k

I'm used to downmix it to mono, since it's only speech, stereo is not neededbut then you loose that whole "talking back and forth" feeling on those commentary tracks where there's 2 or more commentators.:eek:

[raistlin2k]

Well, by downmixing with -d switch this left and right between to commentators gets lost, but both of them are still there, in the center speaker, where speech should always come from, so where's the real problem? I think twice encoding speed and less space taken on CD are bit more important than the discussion-feeling.

Anyway, you are right, but since ogg isn't quite a fast encoder, i prefer downmixing way.

@DSPGURU
Thanks for all your work on BeSweet!

Raist

[Sven Bent]

if you audio encoder containsa a decent joint steros feature/channel coupling which remoced redundacy in the channels there should be very little size difference.

[raistlin2k]

Yes, thats true. Especially at low bitrates - or quality settings - a mono stream has not half the size of a stereo stream.

In numbers:
Smallest possible ( q 0 ) stereo has 58 kbit, but mono has 39, and not 29, being half of 58.

BUT the big advantage of mono for commentary is not really the resulting filesize, but the speed. Since Besweet is not able of creating mono-streams at the moment I made a test with azid and oggenc in 2 steps. Result: azid is 50% faster, oggenc even 100 %, means twice the speed than for encoding the same commentary in (useless) stereo.

Raist

[raistlin2k]

:scared:

Attention!

the mono feature would be nice, but I found out that it is useless at the moment, because when I mux 2 stereo streams,1 monostream & 1 videostream in an ogm, the playbackspeed for the monostream is double-speed (sounds funny:) )

I don't know if this is a problem of muxing or playback, but it makes usage of mono for commentary useless :(

Raist

[DSPguru]

probably a bug with OggDS. should be easy for Tobias to fix.

[raistlin2k]

Well, so we'll wait for Tobias' fix.

Will you add the -o switch to besweet after the fix?
Last time you told me that you'll consider it.

Since I'm actually starting to write my XCDit manual, I would like to know this so I can finish the manual and just wait for the fixes in OggDS, Besweet & the error-correction in XCD.

Thanks
Raist

P.S. Is Speex available for Windows? Have a link?