ogg/opus instead of older ogg/vorbis for compressed soundfonts?

• Jun 18, 2016 - 13:35

opus is a newer audio codec with better bitrate-quality ratios/tradeoff


maybe use it for sf3 or similar compressed sound-font file format?


Better bitrate???

This sentence in Notes in the Wikipedia article is extremely worrying for using it for music.......

Opus cuts audio above 20 kHz, the generally accepted upper limit of the human hearing range.

In reply to by ChurchOrganist

The OP did not say "better bitrate" he said "better bitrate-quality ratios/tradeoff". Different thing.

And even for music, you never need higher frequencies than 20kHz. Your ears cannot hear it, and
your amplification system can probably not handle it.

Here is a good explanation of this:

This is a more visual explanation of digital audio

In reply to by mirabilos

Sure thing. I have no idea if Opus would be suitable for high quality samples or not. I was just pointing out that
supporting audio above 20kHz is a good thing. For samples and for any other kind of audio. We cannot hear it, your stereo/phone/whatever will not support it, and in the worst case would add weird audible artifacts if subjected to such audio signals. And it wastes bits, mostly in terms of requiring a unnecessarily high sample rate (higher than the 44.1kHz of CDs). For the details, see the links above.

I'm sure quality reduction is expected with any lossy compression (no different to now?), but I'm sure it'll be done reasonably. Won't the uncompressed one be available anyway for more serious purposes?

Because Fluid is apparently too big size-wise for the app, could Opus be the solution?

I don't know if SF3 can handle it, but is there documentation for this file format? Generally speaking, information on Wikipedia would be good (it's mentioned here, but it might be best to update this).

As far as I’m informed, Opus does not target music outside of phone or “elevator music” and focuses on the low bitrates, so Vorbis is the better choice, in general, for soundfont samples.

I’d love to see one with embedded FLAC (is there Ogg FLAC?) though, which would preserve editing capability, as well as retain detail for when the samples are still put through processing (attack, velocity, perhaps even transposing to a different concert pitch, and whatnot) and then mixed, you want, in general, the source material to be as good as possible. It also shouldn’t be that much larger than the Vorbis version, and the latter could be generated (at compile time) from the former, which saves space in the distribution as well.

Do you still have an unanswered question? Please log in first to post your question.