Favorite lossy audio format?

>>59815698
FLAC

>>59815706
>lossy
OGG Vorbis

Whatevr handles my music properly. It's not like having preferences in format choice will get you anywhere nowadays.

>>59815839
>not Opus

>>59815839
Vorbis is for earcux

File: 1636222.jpg (6KB, 150x150px) Image search: [Google]

6KB, 150x150px

>>59815698
Opus is the true patrician choice.

mp3 a bloated legacy shit

>>59815698
I like aac.

opus with 96 kbit/s for the mobile phone
flac for everything else

>>59816033
How good does it sound with 96kbit? I mean, comparing it to mp3 at x bitrate

.m4a :^)

>>59816054
near-transparent

opus 128kbps vbr for roughly mp3 v0 quality at half the file size

wiki.hydrogenaud.io/index.php?title=Opus#Music_encoding_quality

>>59816088
Thanks good sir. Wanted to convert my cds to my phone and wanted to try a format with good quality/size ratio.

>>59816054
More than good for me - since I use 5€ headphones on the road.
I choose this bitrate because I read on multiple occasions that 96kbit/s opus is better than 128kbit/s mp3 (which I used before).
But I think that's accurate:
>>59816088

>>59815698
"Create AAC copy" in iTunes. BTFO

>>59815933
what tool can I use to convert my mp3 library to opus?

>>59815698
MP3 VBR

>>59816409
lossless > lossy = ok
lossy > lossy = shit

>>59815923
>>59815933
>>59816033
>>59816088
>>59816145
>>59816190
>cucked to 48kHz

>>59816409
fre:ac

>>59816445
what tool can I use to convert my FLAC library to opus?

>>59815698
JPEG

http://www.cs.huji.ac.il/~springer/DigitalNeedle/index.html

>>59816458
fre:ac

>>59816458
import os
import subprocess

SOURCEDIR = "flac/"
DESTINATIONDIR = "opus/"

lengthSOURCEDIR = len(SOURCEDIR)

for root, dirs, files in os.walk(SOURCEDIR, topdown=True):
for name in dirs:
try:
os.mkdir(DESTINATIONDIR+os.path.join(root, name)[lengthSOURCEDIR:])
except FileExistsError:
pass
for name in files:
if not os.path.isfile(DESTINATIONDIR+os.path.join(root, name)[lengthSOURCEDIR:-4]+"opus"):
subprocess.call(["opusenc", "--bitrate", "96", os.path.join(root, name), DESTINATIONDIR+os.path.join(root, name)[lengthSOURCEDIR:-4]+"opus"])

>>59815698
OGG

>>59816445
>worrying about sampling rate at a lossy format

File: test.jpg (2MB, 2048x2600px) Image search: [Google]

2MB, 2048x2600px

vp8

ffmpeg -i test.webm -f rawvideo - | mpv --demuxer=rawaudio --demuxer-rawaudio-format=u8 --demuxer-rawaudio-rate=32000 -

>>59816508
Thanks pajeet

for i in *.flac; do opusenc --bitrate 96 "$i" "${i%%.flac}.opus"; done

File: test.webm (3MB, 640x480px) Image search: [Google]

3MB, 640x480px

>>59816832
wrong file

OGG

>>59816746
>i use a lossy format which on top of it being lossy forces a resampling as well

>>59816409
dbpoweramp

>>59816850
That leaves me with both .flac and .opus in the same directory.
All but my last line is for creating (if not already created) a folder structure similar to the one where the flacs are stored.

>>59817361
You can copy the source dir and

 find . -name '*.flac' -exec rm {} \;

after

opus.
If you have to use legacy, try aac.

>>59817421
That's true. I guess there is even a faster way by moving just the opus files instead of copying everything and deleting the flacs afterwards.

I can add one new album in the flac directory, run the script and have only this new album converted in my opus directory. Works like a charm.

For one time conversion, your way is way better. But for me, Python is the way to go. I guess its a good thing that I never started to do anything IT related professionally.

>>59816033
Is there an android player that can play opus? I have poweramp but my music refuses to appear.

>>59818348
I use Neutron for that purpose.

>>59816458
what the fuck is wrong with you

ogg+aotuv encoder

File: 1483118158661.png (55KB, 217x190px) Image search: [Google]

55KB, 217x190px

>>59815698
Opus

>>59815698
Favorite lossful format?

>>59815839
Ogg is not an acronym. It's a name.

>>59816514
>>59816983
Ogg is a container, not a format.

>>59815698
What does lossy mean?

>>59821563
Not same quality as the source. Lossy is usually smaller than lossless because it discards some information.

MP3. It just werks.

File: Screenshot_2017-04-09-22-46-40.png (358KB, 1440x2560px) Image search: [Google]

358KB, 1440x2560px

I use OGG. I think like 400kbps or something in sound converter.

>>59816409
opus-tools

i exclusively listen to .rma rips

>>59821827
You're spoonfeeding a literal normie retard. Why?

>>59816832
Wow, that doesn't sound nearly as bad as I thought it would.

>>59816832
>>59816859
Are you a wizard?

AAC for small battery footprint
OPUS for anything else

>>59815933
>mp3 a bloated legacy shit
This. I have no idea how it can have survived this long considering a lot of music players from like 2005 supports at least Vorbis.

Opus, period.

Opus. I use AAC more for audio track of video, though, because of muh Blu-ray player.

AAC.

Opus has a pretty stupid encoder that will try to hit the targeted bitrate no matter the complexity.

>>59821856
>400kbps
son, you goof'd. ogg with the -q5 is transparent to normal, non-golden ears. if you still feel that you need some extra placebo -q6 is well enough. if you want the maximum placebo, go lossless.
>also, ditch ogg and join the opus master race

>>59816508
>>59816850

you vs the guy she tells you not to worry about

>>59823922
If you're already using vorbis there's not really enough of a difference between it and opus to justify a switch, especially if you have a large library to reconvert.

>>59823809
Huh. Try some harpsichord track and opusenc gives 1.5x bitrate or more.

>>59824759
320k gives me 296k on a 330k FLAC.

>>59815698
AAC. I replaced my mp3 with m4a (aac) from iTunes "free" site

>>59824814
That track is not that hard to encode for opus, then. --bitrates 96 gives 80 to 133 for those samples.

http://listening-test.coresv.net/results.htm

>>59823409
i have 3 years of training left

>>59824898
It gives 296k on a 330k FLAC.
qaac q127 is 166k.

The Opus encoder is retarded.

>>59824959
Is that piano solo or similar tonal music? If so, it's kind not that easy signal for Opus because of its small overlapping MDCT. It's more efficient than AAC for oth music.

>>59824996
It's an OVA from 1991 with pretty much nothing above 16KHz.

>>59825027
Mono? Use half bitrate then.

>>59825053
No.
Oh, that Opus doesn't know the channel configuration makes it doubly stupid.

File: 1489226236855.gif (2MB, 540x501px) Image search: [Google]

2MB, 540x501px

>>59816458
I use find, parallel, and opusenc in a script.

>>59825148
You need to upload 30 sec sample and give it to devs. Oh, and it encodes mono to multichannel fine although it doesn't support uncommon mappings such as 6.1.

>>59825205
I doubt it's unknown that it targets the bitrate, given it will target it regardless of the channel number.

qaac q127
>mono: 160k
>2.0: 320k
>5.1 768k
>7.1 1000k

opus 320k
>mono: 320k
>2.0: 320k
>5.1: 320k
>7.1: 320k

>>59825390
It's fine if the quality is constant within each channel mapping.

>>59825535
It may be per recording, but the bitrate doesn't have much of any correlation to the complexity.

>>59825581
Opus is rather different from other encoders. Give this simple piano piece to opusenc and you'd get rather high bitrate.

http://www.mediafire.com/file/25gt0pej7x38m3b/mazurka.flac

>>59825660
yeesh, 487kbps vs qaac q127 241.

AAC/MP3, because it's hardware decoded on my phone/player. Playing opus kills the battery faster and causes the phone to get warm.

>>59825708
Opus gives really noisy file on 96kbps if you force --hard-cbr for that sample. I'm quite fine that default VBR gives proper quality although it bloats bitrate for killer samples like that.

http://www.mediafire.com/file/w70buuaeiseszol/96cbr.opus

After trying AAC, AAC-LC, Opus, AAC-HE, MP3, DTS, and even AC-3: Opus is the clear winner.

The artifacts it does produce are not as violent as the artifacts found in most of the codecs and achieves audible transparency at 128 vbr for me though I use 192 vbr for good measure.