Will begin the process of porting bits of sndkit and graforge to my Rust library , as part of December Adventure.
No promises about daily consistency, since I'm living a dual life over at @patchlore.
I log hours for all my work now, so I'll eventually be tracking progress for this stuff here:
Spoiler alert: My GP didn't go back to bulk-billing https://phonakins.com/2023/11/29/spoiler-alert-my-gp-didnt-go-back-to-bulk-billing/ #Autism, #Disability, #Dsp, #Medicare
• The Scientist and Engineer's Guide to Digital Signal Processing—Smith (basic introduction)
• Understanding Digital Signal Processing—Lyons (advanced introduction)
• Digital Signal Processing—Oppenheim (comprehensive classic; also see https://www.youtube.com/watch?v=rkvEM5Y3N60&list=PL8157CA8884571BA2)
• An Algorithm for the Machine Calculation of Complex Fourier Series—Cooley (the origin of the species, as it were)
• MATLAB Digital Signal Processing System Toolbox—MathWorks (for basic lab work)
• Digital Signal Processing using Arm Cortex-M based Microcontrollers: Theory and Practice— Ünsalan (for advanced lab work)
• Digital Audio Signal Processing—Zölzer (for practical knowledge)
• Wavelets and Filterbanks—Strang (more EE perspective)
• Digital Filters—Hamming (another classic by a pillar of DSP)
• Fourier Series—Tolstov (comprehensive theory)
Aberrant DSP Lofi Oddity (Degradation Toy Box) v1.0 x64 VST3 AU AAX WiN MAC [FREE]
Vladyslav Voinov PeakEater v0.8.2 x64 VST3 AU LV2 CLAP LiNUX WiN MAC [FREE]
#AU #CLAP #LV2 #VST3 #Analyzer #artifacts #AU #boost #Change #CLAP #Clipper #DAW #DSP #easy #effect #filter #free #fun #Future #Gain #GitHub #Hard #Host #info #Linux #LV2 #MacOS #Minimal #other #Peak #RMS #Time #VladyslavVoinov #volume #VST3 #x32 #x64
Valhalla DSP Valhalla Supermassive v3.0.0 x64 VST VST3 AU AAX Windows MacO [FREE]
#VSTVST3AAXAU #2023 #AAX #Andromeda #AU #bass #BigSur #Dark #delay #DSP #echo #Effects #EQ #feedback #free #Hydra #Intel #M1 #MacOS #mind #Modulation #Morph #music #Network #notes #other #Patterns #Presets #Reaper #reverb #space #Spring #Stereo #Strange #Strings #Synthesiz
I recently found on Cornell #arXive a new pre-print (2023) on #RNN and #LSTM by Alex Sherstinsky of MIT. Through the years, I've read numerous papers on RNNs, starting with Rumelhart's 1986 paper. But this one is, by far, the most detailed tutorial not only on RNNs but also on LSTMs.
The complete derivations of both forward (inference) and backward (training) passes of the learning algorithm use only basic calculus and matrix algebra, drawing intuitive analogies to digital signal processing #DSP. And the equations are complete and detailed enough to be implemented by the student, directly in software. In my opinion, every undergrad EE and CS studying #DeepLearning #NeuralNetworks should read this superb introduction.
Well, that was a fast turnaround (MY NEW BATHTUB IS PURPLE) https://phonakins.com/2023/11/17/well-that-was-a-fast-turnaround-my-new-bathtub-is-purple/ #Christmas, #Dsp, #Family, #Housing
Onkyo ha presentato una settimana fa il ricevitore audio video Onkyo TX-SR3100. Il nuovo Onkyo TX-SR3100 è progettato per essere un ricevitore audio-video (AVR) a 5.2 canali, facile da usare, prezzo conveniente ($399) e ricco di funzionalità e con una qualità finale vicina a a quella dei ricevitori di fascia alta.
In un sistema audio 5.2 canali:
Il numero 5
Demo (images are only processed locally on your own device):
This is by far not the fastest way to compute this, but it explains the principle approach more clearly. Almost exactly 13 years ago (to the day), I used the same with Java and my toxiclibs to transform images, driving altogether 768 granular oscillators to create super complex, constantly morphing drone sounds. Some selected audio snippets & images are here:
 It's called "waveform scope" in Adobe Premiere, even though the term is misleading...
Karanyi Sounds Electric Keys v1.0.34 x64 VST3 AU RETAIL WiN MAC [Free For Limited Time]
#AU #VST3 #VSTI3 #ADSR #ambient #AU #Classic #DAW #DSP #easy #effect #Effects #EQ #FM #free #fx #Gain #ImmerseYourself #KaranyiSounds #Keys #layer #limited #Maschine #MIDI #Mix #music #Piano #Presets #Random #Retro #reverb #SoundDesign #space #Synthesizer #Time #transform #
Hey! If you know how to write clean #python, there's still some GNU Radio #hacktoberfest low-hanging fruit issues that could use your help. Not all of these require any #DSP knowledge.
Many of them are just a place where you can showcase how well you can write good Python!
For example, @gnuradio has configuration system, and it's seriously undertested. Add good test coverage!
J’utilise beaucoup d’approximations de fonctions (trigo, exp/log, …) dans mon code. Elles sont plus rapide à évaluer ce qui est important en traitement audio temps réel.
Une caractéristique importante de ces approx est leur précision, résultant d’un compromis avec la vitesse.
J’en ai toute une collection que je tiens à jour, avec tests de vitesse et précision sur diverses architectures.
Je me suis dit que ce serait une bonne idée d’en tracer les graph d’erreurs détaillés.
First, the government.
#AKP: 242 in favor, 1 against, 20 absent
#MHP: 49 in favor, 1 invalid
#YRP: 3 in favor, 2 absent,
#HÜDAPAR: 4 absent
#DSP: 1 in favor
Very surprising to see one against vote from the ruling AKP. It was from Burdur MP, Mustafa Oğuz. I don't know anything about him, he was probably just confused. Two things to underline. First, there is not a single absent MP from the far-right nationalist MHP. Second, all four MPs from #Kurdish #Islamist Hüdapar were absent.
This is our reality.
This the choice countless Australians are having to face each & every day.
Assuming they have the privilege to have a roof over their head.
Poverty is a political choice.
Give us the tools and the support to get out of poverty.
An article by Chips and Cheese covers Qualcomm's VLIW Hexagon Processor: https://chipsandcheese.com/2023/10/04/qualcomms-hexagon-dsp-and-now-npu/
If you are an #unijena student and this sparked your interest, consider taking our #hpc lecture which provides a more gentle introduction to low-level Hexagon and System-on-Chip (SoC) programming in general: https://scalable.uni-jena.de/opt/hpc/chapters/assignment_soc.html
Given measured event times
t0 = 00:00
t1 = 01:59
t2 = 02:57
t3 = 04:50
t4 = 08:24
a priori: period is roughly 1h and is changing length slowly
Eyeballing this data, I can see that t1->t2 must be one period of 58m. Then I can tell that t0->t1 is a 60 and a 59 and t2->t3 is 57+56. So I know the period is changing 1m/cycle.
Now do it with noisy data, but answer is approximate
I've reimplemented the diode clipper and bassman tone stack using wave digital filters and an impulse response convolver from scratch in #cpp for my final year project. feels really good to see matching results compared to chowdsp's byod
My hands are shaking as I open this vintage Led Zeppelin boxed set for LONO Audio to demo our remastering technology on these priceless classics. (No, they are not a signed customer; it's just for a sales pitch. But still. Wow.) #classicrock #music #musictech #mastering #remastering #dsp #signalprocessing
I based my design on what I wrote a long time ago for the Ohm Studio built-in FXs.
It sounds nice at the moment but I see a lot of sonic improvements to do.
It runs like breeze on a desktop CPU but loads significantly (22 % in a mono setup) the poor #RaspberryPi4 Cortex A72 when all features are activated and modulated. Time to start aggresive optimisations!
This is how our government ensure people in poverty stay there.
Whilst we slowly starve, struggle to find a roof over our heads they boast about a $22 billion surplus whilst giving further tax breaks to the wealthy.
They simply don’t care about the disabled, the downtrodden, the hard working members of society.
ICYMI our politicians had been hard at work to ensure the funds of that $22 billion surplus never ended up where they were needed.
Furthermore, they made sure it wouldn’t reach the pockets of the most needy and vulnerable.
Just updated the synth & sequencer of the most recent #HowToThing example to be much more musical, added more params and switched to a faster oscillator. Now you can generate up to 5 mins of audio, choose note length, number of octaves and probability...
New audio example in the reply...
#HowToThing #015 — A special one: Creating a polyphonic & multitimbral synth with stochastic sequencer to generate offline audio (*not* WebAudio!), rendered via fibers (co-routines) and exporting the result as WAV file. Each synth voice uses a randomized config & FX pipeline for each note played (osc → adsr → SVF → filter delay line). Generated audio is attached.
Key packages used:
- https://thi.ng/dsp: Signal generators & effects
- https://thi.ng/dsp-io-wav: WAV file format export
- https://thi.ng/fibers: Time-sliced audio processing
- https://thi.ng/rdom: Reactive UI construction/updates
This project is also available as new example #130 in the thi.ng/umbrella monorepo (originally based on a workshop exercise @ University of Applied Sciences, Augsburg, Nov 2022)
(Code screenshots in reply this time... ✌️)
Virtual Vocaid ROMpler Plugin by SampleScience
#2023 #ADSR #AU #chorus #delay #distortion #DSP #Effects #filter #highpass #info #LFO #Library #lowpass #MacOS #Modulation #Multi #pan #Pitch #polyphonic #reverb #Rompler #SampleScience #synthesis #Voice #VSTI #VSTI3 #Windows #Words
It's a good thing I brought the #guitar from yesterday home already or I'd be playing it right now instead of working (and now I'm talking about it anyway)
#hyperfocus is bad enough, but I'm also much happier when I'm hyperfocusing
I can't even tell myself that One Day This Will Be Professionally Useful like some other recent focus points (#linearalgebra #dsp #fft #python #kalmanfilter #orbitalmechanics #geometricalgebra #electronics)
Así que, la expansión estéreo, convolución y ecualizador paramétrico sólo son disfrutables con audífonos...
Te invito a derretirte el cerebro escuchando música así de vez en cuando
Es lo que los esnobs y demás mamadores sueles llamar como: sonido 8D (ja!)
Yo :bc_heart: #JamesDSP!
#Audio #DSP #Android #Convolution #EQ #Music
Good Morning & welcome to #HowToThing #004 — Painlessly creating text-based plots to debug & visualize sequential data in a REPL-driven workflow. Here we're using https://thi.ng/dsp signal generators, but _any_ numeric array or iterable will work. Visualization via https://thi.ng/text-canvas (new versions of both packages have just been released...)
This is where I walked to, and sat, after getting the news in writing that the literally crippling illness I have, that makes me unable to study or work more than an hour a day, completely disqualifies me from #NDIS and #DSP.
My family *actually would* be better off if I was dead, because they're the ones forced to provide for me.
My gov acknowledges I cannot work - and then requires it. My family have told me they're at breaking point.
So... I dunno. Fuck me.
First meaningful plot for my project. It's a WDF model of a parallel diode clipper (like those in many distortion pedals).
I can adjust the number of forward or reverse diodes to change the amount of positive or negative clipping.
Now I need to try build a vacuum tube model from this. I have a couple papers for vacuum tube models already but I've been chewing through them the past 3 weeks with little progress. I'm hoping this is a step forward.
Soundiron Rainsong 2.0 KONTAKT [FREE]
#AAX #ambience #ambient #AU #City #Classic #delay #Digital #distortion #drive #DSP #easy #effect #Effects #EQ #Fade #filter #Fire #flanger #free #fx #Hard #Hardware #Impulse #Keys #Kontakt #KontaktPlayer #layer #LFO #Library #lowpass #Mix #Modular #Modulation #NativeInstruments #NKS #notes #Online #Pads #Patches #phaser #Pitch #player #plucks #Presets #Rain #RAW #rev
🎈✨ After 2.5 years of development, I'm excited to have released Facet v1.0.0! 🎉🎂
Facet is a live coding environment for algorithmic composition and synthesis.
With a code editor in the browser and a pair of NodeJS servers, it can generate, transform, and sequence audio, MIDI, and OSC data.
There are over 150 methods which can be combined in numerous ways to create unique DSP algorithms. These methods create and modify arrays of floating-point numbers which become lossless audio files or MIDI/OSC patterns.
In order to play back synthesized audio patterns, Facet uses a Max abstraction as a "receiver" for up to 4 channels of audio, and from Max it can go anywhere. The Max abstraction also pipes OSC out of its fifth outlet, making it possible to "live code" a prepared Max patch.
Also included in the repo are two Max for Live devices (stereo and 4 channel version), allowing you to send audio directly into Ableton Live tracks. With this setup, you can also create MIDI CC mappings to control Live device parameters.
If working with MIDI or OSC only, Facet can run as a standalone system, making it possible to run on a Raspberry Pi or computers with less CPU resources.
Facet runs on MacOS, Windows, and Linux.
I do enjoy spam threats from the government.
Yes, Mr Varadkar, I would like a job or some contract work too. That would be nice! I didn't expect to be unemployed this long either! 🙃
I applied for a job yesterday and have actively been trying to find even short term contract work! I would rather not be unemployed...
Howdy tooters, any ideas how to compress floyd-steinberg dithered b/w images? The result doesn't need to be 1:1 with the original, it just has to look good. Traditional lossless compressors struggle with these. I have some ideas that I'm yet to try, but would love some more.
#compression #algorithm #algorithms #dsp #askfedi
@acousticmirror @daviwil #faust is so mind blowing tbh. its just so good, you can pick up any paper and implement it in minutes just from the diagrams, without even actually reading the paper. and the workflow of the diagrams hot reload is super satisfying. its honestly the best example of a the power #DSL(s) can offer. if you're already familiar with zero-delay #DSP you can pick it up in 5 minutes, and port things that require a small mountain of #csound code to like, 10 lines of faust in 15 minutes. its bafflingly good, and extremely declarative. if you understand a design — or hell, better yet, even if you *don't* understand it but have it ready at hand — you can implement it at the speed of reading.
last I was using it about 2 years ago it suffered from not having a proper fft story, which is certainly a significant area to not cover, but spectral processing is kinda nerdy anyway lol. what Faust offers is the best possible set of tools for getting deep and dirty with the most fundamental object of interest to the real dub heads — infinite impulse response filters, and the precise control over non-linearities they offer as elaborated in the filter theory of Vadim Zavalishin.
@black_intellect I'm sorry, but the way *I* read this, it's DSP's choice to be a vendor for only one company: Amazon. Amazon does not need to be a part of their negotiations. #DSP has a contract with the drivers already. What part of business do these drivers not understand?
If you want to work FOR #amazon, work FOR them, not a vendor!
As it is, I feel nothing for the drivers. I work FOR #uber. We have walked off the job to protest before. We got changes.
Chowdhurry DSP Chow Centaur v1.4.0 x64 x86 VST VST3 AU LV2 STANDALONE WiN MAC LiNUX [FREE]
Last batch of the WhateverDSP open source DSP pedals ready to ship out to prototype testers!
Interested in making one yourself? Check out the git repository! https://github.com/NuclearLighthouseStudios/FXDSP
We also have a discord now (yeah, I know…), so why not say hi if you're interested. https://discord.gg/8hGEM5nK5X
Also gonna do another batch of preassembled ones soon if you don't want to build your own!
Chowdhurry DSP Chow Kick v1.2.0 x64 x86 VSTi VSTi3 AU LV2 CLAP STANDALONE WiNDOWS MACOS LiNUX [FREE]
Here's what that means: when an advertiser wants to place an ad, it contracts with the "#DemandSidePlatform" (#DSP) to seek out a chance to put an ad in front of a user based on nonconsensually gathered surveillance data about a potential customer.
The DSP contacts an ad-exchange - a marketplace where advertisers bid against each other to cram their ads into the eyeballs of a user based on surveillance data matches.
As programmers, we often read about about CISC and RISC cores, but Sega’s SVP chip used a #DSP core.
Digital Signal Processing (DSP) algorithms require a large number of mathematical operations to be performed quickly and repeatedly on a series of data samples.
Based on a #DSP core, the SVP chip allows the Genesis to render real-time polygons and provides an "axis transformation unit" to handle scaling & rotation.
Virtua Racing (1994) was the only game released with this chip, and the only Genesis cartridge with any enhancement chip!
The Sega Virtua Processor (SVP) was a 23 Mhz custom-designed Genesis-compatible #DSP chip to enhance graphics and sound.
The SVP chip basically served as an extra processor, allowing games to push faster polygons than what was possible with the vanilla Genesis hardware.
A Group of #AmazonDrivers Just Joined One of the Biggest #Unions in the US
Drivers in California have joined the International #BrotherhoodOfTeamsters, in one of the first driver-specific unionization efforts at #Amazon.
The group of 84 delivery drivers in Palmdale, California work at Amazon’s DAX8 delivery station. They’re not #Amazon #employees, but rather employees of an #AmazonDeliveryServicePartner (#DSP) called Battle-Tested Strategies.
I'm working on an experiment involving embedding audio samples
I'm performing feature extraction using "Mel-frequency cepstral coefficients" (I don't understand very well but thankfully there are good libraries). That gives vectors which can be embedded down to 2D after a bit of cleanup and post-processing
I'm thinking about possibilities for building a sample browser based on this, letting you find similar samples and explore your library visually
Wavesfactory Typewriter v1.1.0 KONTAKT [FREE]
Typewriter is a Kontakt sample instrument featuring a Royal vintage typewriter.
Meticulously sampled at Calmaestudis recording studios using 4 mic positions: close s
Library to provide UGens and utils for the excellent Incudine Music/DSP system.
Currently its just providing 'analog' bandlimited saw, square, pulse and triangle waves. More planned.
The API is early, and will probably change. Zero documentation. Use at your own risk, but I am interested in feedback.
Next plans are to work on documenting the existing filters in Incudine in a tutorial, before filling in some possible gaps (VirtualAnalog filters for sure).
Hi, my name is Mads and I am an electronic music composer/producer and software developer. I enjoy making experimental electronic music and #techno using DIY software and hardware. I work at Torso Electronics by day, making electronic music instruments for the future.
I am also a massive #tea nerd.
I publicize some of my code/hardware here
Just heard back from GRIT's owner, who tells me the problems other #DSP owners have had was not her experience, and that she chose to close her business to "focus on other tasks."
I have updated the story with her comments.
#GeorgeUpjohn, a former pilot in Sydney with brain cancer has been repeatedly rejected for the #DSP because his condition is steadily deteriorating. He's had to sell his possessions and borrowing increasingly large amounts of money from family to stay alive: https://www.theguardian.com/australia-news/2021/oct/28/turned-away-how-strict-government-rules-denied-young-pilot-with-brain-cancer-the-disability-pension
What happens to George when the same state that has refused to offer him the supports to live enthusiastically informs him that he qualifies to die?
The Australian state pays a *significant* amount of money to support me and people like me. What happens when they figure out instead of paying our #NDIS plans every year they can just pay a one-time cost to off us?
If this scares you, I'm glad. It scares me.
#ActuallyAutistic people in #Australia already die 17 years sooner than other Australians. I know I'll probably get less time than my predecessors: https://onlinelibrary.wiley.com/doi/10.1002/aur.2086
(This pales compared to Autistics in the #US, whose life expectancy is *36*, or even #Sweden, the only other country where we have data, where it's only 54): https://ajph.aphapublications.org/doi/abs/10.2105/AJPH.2017.303696?journalCode=ajph
Many disabilities are similar: we are among the lowest-valued people in the world.
I want to thank
for her incredible reporting on #MAiD over the course of years: most things I learned about MAiD were either from her, or via people I found from her page.
For the rest I've been advocating on these issues & studying policy for about a decade.
The pressures disabled people face in #Australia are not altogether very different from those of us in #Canada. Nearly half of disabled Australians live in poverty, over twice the OECD average and 2.5 times the Australian average: https://www.aruma.com.au/about-us/about-disability/disability-statistics/
We have 2x the unemployment rate & are more at risk of domestic violence, sexual assault, & mental illness. 100,000s of us sit on pensions that are half the Henderson poverty level, and less than 1% of rentals are even affordable to a person on the #DSP: https://www.miragenews.com/hearing-to-inquire-into-experiences-of-poverty-844466/
Our government has steadfastly rolled back pandemic protections over the last year: they know this plan will mass-kill disabled people, and create new disabled people, and they don't care.
So what’s going to happen when the state, refusing to offer enough resources to live, starts offering death on demand?
Many people don’t even qualify for disability services: non-citizens, except for refugees & asylum-seekers, can't get #Medicare (health care).
Non-citizens don't qualify for the #NDIS (support services) at all, regardless of disability severity.
Even citizens can't get the DSP (pension/money) without 10 years' residency and their condition being "fully treated" & "stabilised."
What does this mean?
#Introduction and some made-up hashtags of my interests, which I will get wrong and then edit:
* #digitalart #krita #digitalpainting #blender #photography #landscapes #architecture
* #sff #writer
* #minecraft #tableTopGames #boardgames
* #synthesizers #music #musicComposition #surgeSynth #surgeSynthTeam #dsp
* #theoreticalPhysics #philosophy #maths
I might occasionally post my creative explorations.
Another #introduction since I moved servers.
I'm into #programming both professionally and as a hobby, lately taking sidesteps into #dsp and #electronics and in love with #rustlang. Currently working on something on my free time that might become both a #clap / #vstplugin and #eurorack module if stars align.
#electrical #engineer from #Nashville, TN. #exvangelical hanging with #ACNA #Anglican. Playing catch-up on #theology #ethics and #hermeneutics. The #Bible #scripture is textually fascinating! Ask me about #GospelOfMatthew and the #SermonOnTheMount. #ecumenism #orthopraxy
Christian love demands #science!
I guess I should write an #introduction.
I've been using the internet since 1992 and contributing to #Wikipedia since 2001. I haven't used Twitter since early 2021 because I decided it was unhealthy, and I used Mastodon a fair bit that year, but then lost my account through, apparently, inactivity, which I didn't realize was even a thing.
I guess I should put some kind of tag cloud here of stuff I'm interested in? Maybe #physics, #Linux, #Debian, #manufacturing, #DSP, #Python, #Numpy, #math (more #algebra than #analysis), #raytracing, #commonsbasedpeerproduction, #creativecommons, #piracy, #scihub, #libgen, #emergence, #solarpunk, #history (especially #Hellenistic and #Mohist), #UX, #HCI, #controltheory, #lifesupport, and #infosec.
I'm a retired software engineer in #PDX. I worked on #LLVM and other #compilers. Earlier, I did #DSP and #embedded computing. I was trained in #mathematics originally. I like #electronics and #microcontrollers.
I started a #woodworking hobby. I prefer to work with hand tools and go slow. I'm not starting a table factory to feed my family, so it's okay.