Hi.
Since I rely on accessibility and haven't found a "DJing tool" for Linux
which doesn't require a graphical interfaces...
I used to do offline mixes with a bunch of sox instances piped into each
other, orchestrated with a shell script...
Which is rather tedious, so...
I recently came back to this and decided it is a great opportunity
for having a tool vibe-coded.
For playing with vibe-coding, and actually solving a problem I always had.
So I had GPT-5(.1) scratch my itch.
And out came
https://github.com/mlang/clmix
It has a built-in player + metronome for working out bpm, offset
of the main beat grid and start/end cue points.
Seeking works with bars, so you dont have to fiddle with time.
Once you worked out a track by ear, you can do a "snap-to-grid"
using aubio as a beat/onset detector to refine the offset if necessary.
Once you are happy with the track timing, you save it to your DB.
If you have a number of tracks in your DB, you then can
generate a (shuffled) mix. Selection of tracks from the DB
uses a tag expression language, so you can do things like
clmix music.json --random "dnb | jungle" --export beats.wav
clmix will determine the mean bpm, bring all track to the same
tempo and align cue points for you.
Thats basically its main use case for me.
I use mpd with crossfade a lot for playing music at home,
and kind of wanted something that actually can generate good crossfades.
So when I want a "club mix" at home, I now simply
generate it from my favourites I added to the clmix db.
Nothing special, just a tool to get a particular job done,
for a particular user group (command-line junkies).
Vibe-coding was fun.
From idea to first executable code was just a few hours.
However, I have a particular taste when it comes to C++,
so after a while of being amazed that much apparently correct code
was written by the AI, I started to play the "let me fix this" game.
Ended up doing a lot of bikeshedding. OTOH, I still feel
that kind of ping-pong was necessary to keep the code half-way decent.
If you let the AI do its thing without supervision, the code
ends up pretty unreadable after a while.
In any case, I make the fact this is vibe coded clear because
I do definitely not claim the fame. While I love to play
with programming, this is neither my dayjob nor do I have a lot of
(recent) practice. So I likely wouldn't have written this
if I didn't have some help. So with all the war around AI use
on the net, I can say this was a nice outcome for me.
Nobodies job was taken away, and I got something nice for maybe $30 of
API costs.
I got what I wanted, and I didn't have to waste a lot of time on it.
In any case, if this is helpful to anyone, enjoy.
--
CYa,
⡍⠁⠗⠊⠕
Hi,
I am trying to locate the firmware for RME HDSP Multi- and Digiface on
Debian stable, as referenced on https://wiki.debian.org/Firmware/List
but it seems none of the packages firmware-linux-nonfree
firmware-linux-free nor alsa-firmware-loaders
do provide these files (anymore). Is this intenional, and is there
another official source for
multiface_firmware.bin
multiface_firmware_rev11.bin
digiface_firmware.bin
digiface_firmware_rev11.bin
?
Thanks for all pointers,
best, Peter
Loopino — New Release: Faster Workflow, Deeper Control, More
Sound-Shaping Power
The latest Loopino update brings a major boost in workflow, sampling
flexibility, and creative sound design tools. Built for fast
experimentation and musical expression, Loopino now makes it easier than
ever to load, capture, shape, and export unique audio textures.
Effortlessly drag and drop samples or browse your files directly inside
Loopino. Record new material on the fly, trim to precise clip marks, and
refine your sounds with non-destructive fade-outs and integrated pitch
tracking. The Micro Loop Generator has been expanded to let you choose
loop count and duration, turning any waveform into evolving rhythmic
fragments or playable micro-textures.
For deeper shaping, Loopino includes non-destructive wave sharpers
(square & saw), a full ADSR envelope, and a versatile LP/HP ladder
filter with cutoff and resonance control. Phase modulation sources
(sine, triangle, noise, Juno-style), vibrato, tremolo, and root
frequency control open the door to expressive synthesis possibilities.
And with up to 48 voices, Loopino handles dense, layered sounds with ease.
Export your processed samples or micro-loops as WAV files—automatically
tuned to your selected key—and save or load presets to build your own
library of sounds.
Key Features
* Drag-and-drop sample loading
* Integrated file browser
* On-the-fly audio recording
* Trim samples to clip marks
* Non-destructive fade-out processing
* Integrated pitch tracker
* Micro Loop Generator with selectable loop number & duration
* Square & sawtooth wave sharpers (non-destructive)
* Full ADSR envelope
* Preset save/load system
* Export processed samples/loops as WAV in selected key
* LP/HP ladder filter with resonance & cutoff
* Phase modulators: sine, triangle, noise & Juno-style
* Vibrato & tremolo
* Root frequency control
* Up to 48 voices for polyphonic playback
Loopino continues to evolve into a flexible, creative sampler-synth
hybrid—perfect for sound designers, experimental musicians, and anyone
who loves transforming audio into something new.
Availability
* Linux: Standalone application, CLAP plugin, VST2 plugin
* Windows: CLAP plugin, VST2 plugin
Project Page:
https://github.com/brummer10/Loopino
Release Page:
https://github.com/brummer10/Loopino/releases/tag/v0.0.2
Halle,
is there any OS softwre which I can run om ny
webserver to handle communications withon a choir,
looking around I only see commersial ones ?
Regards,
/Karl Hammar
