aide team mailing list archive

[actionparsnip <question694264@xxxxxxxxxxxxxxxxxxxxx>]

Question #694264 on soundconverter in Ubuntu changed:
https://answers.launchpad.net/ubuntu/+source/soundconverter/+question/694264

actionparsnip posted a new comment:
>From Google Translate:

So here I am going to ask a question, I do not know if it is rather a
bug, a problem of the Gstreamer encoder or if it is due to me. Please
don't hesitate to ask me to write in English, there is no problem with
that.

I am using SoundConverter 3.0.2 downloaded from Ubuntu packages.

My operation is quite simple. I am converting flac files (no quality loss) to opus with high quality setting (~ 128 kbps), nothing else (I am not resampling the source file). It turns out that I came across 2 very different cases:
- if the input file (flac) is only music (therefore without a person's voice), the output file (opus) at a bitrate 374 kbps - which is huge not compared to the desired bitrate of ~ 128 kpbs.
- if the input file (flac) is a song (therefore with the voice of a person), the output file (opus) at a bitrate 130 kbps - what is expected.
First observation is that the SoundConverter program puts the voice recognition in automatic mode or not (but does not indicate it in the parameters). Second observation is that whatever the sampling frequency (44.1, 48, 96 kHz), we remain on the same observations (between 2 and 3 times more flow - bitrate - for musical audio).

I was wondering if it was more a problem of the Gstreamer opusenc
encoder or if it could be "the fault" of SoundConverter.

Thank you in advance for your reply.

-- 
You received this question notification because your team Communauté
d'aide d'Ubuntu en Français is an answer contact for Ubuntu.