Well, if anyone ever gets me a copy of SVCTL32.dll / SVCTL64.dll without the timer restrictions, would seriously pay like a hundred bucks for it. But it's never gonna happen. Sigh. That DLL is forever time-locked to something Image Line do in FL Studio to give it proper licensing. So no newer softVoice, no 64-bit SoftVoice build until then. The DLL files at eurpod.com/svfl.zip for those interested. SoftVoice as a company is quite defunct, their website is up yes but as far as a company with filings and public record they have not been around since the mid-2000s. So, not a chance of reaching out and asking for my own licensed copy either. This is all making me very sad, now I've lost all motivation to code. I tried to see if it was returning the last chunks of text in the callback, but nope. SVAudioCallback is either a query function (not a “register callback” function), orSVCTL only enables callbacks after some other registration step (likely via SVRegister / SVCommand), or callback is only for a different output mode (like “send audio to file” or “external device”), and isn’t used for TTS the way we’re invoking it. That's why my hunch here lies with the demo restriction and a timer that is set for each utterance, that when stopped throws away the last chunks of spoken feedback.
Curious if some enterprising hacker could do some editing of the DLL itself if it's not super obfuscated or whatever. Note that I don't have the requisite knowledge to know if this sort of thing is generally possible, but since cracking teams seem to be able to do this kind of thing, I assume it's doable with effort.
I'm surprised there isn't a way you can make your own fake DLL that looks and smells like the real thing, but is really just a surveillance drone so you can find out what FL Studio is doing to register/license the thing.
@jaybird110127 well, FL Studio has speaker. dll. it loads when you use Speech Synthesizer (Windows-only feature in FL). FL (or the Speech Synth plugin/module) calls ZXSpeakInit in speaker.dll. Then, speaker.dll loads/uses svctl (“control” layer), and svctl loads/uses sveng (the actual voice engine). (This is implied by the dependency on SVCTL32.dll and the “3 file” requirement.) speaker.dll then asks SoftVoice to synthesize the text, likely into a PCM buffer. Finally, speaker.dll (maybe via MSACM32.dll) converts/normalizes the audio into a format FL expects. So in theory inspecting details within speaker.dll can give all the clues too, but reading disasmed dll is so not een in my pervue, lining up function names to areas of the DLL, loading it in debuggers to check what it dos in memory, ETC... True about DecTalk though. It's why I was surprised when Blazie used it in BTSpeak, though perhaps there's some other licensing / conversations we don't know of there so I won't assume anything. Still, it's either that or clear they saw low-risk at distributing it more commercially within BTSpeak units. Very interesting. And yet, Apple themselves probably will never touch it, despite it being so in the open in sources even if unofficially, they just know they can't touch it with the historical significance and potential licensing rights for them.
I was about to suggest you contact the registered owner of text2speech.com but of course it's a private registration. Chances are it wasn't always that way, but I have no idea if there's an archive of old whois data anywhere. But of course the fact remains that even if the actual company is dead, somebody, somewhere *is* paying to keep the domain renewed and the website active.
@jaybird110127 oh interesting. Yeah, that's very mysterious, because the site is just mostly a zombie one at this point, with content only archived to be archived. But I do wonder how FL Studio (well, Image line) got to build a new binary around 2014 or so, at least that's the timestamp on the DLL. It looks like even FL Studio 2025 includes it, so they especially are keeping it alive but of course I think probably have full source code or SDK if they had purchased it back when the company was still alive.
As a post about domain names comes across my timeline, I am also reminded, for the second time, by the Dot Radio organization that netmix.radio is up for renewal on February 12. Every time I look at the listener count over there, its usually me, myself, or no one, LOL. But that's totally expected. I just through up an audio-processed stream of 90,000 tracks of randomness with a couple jingles and let rip. So we'll definitely call that a half-fast job. I didn't expect to get much out of the project, and I probably don't want something like that getting too big anyway. So it'll probably not get renewed.
Hmm, I get a problem loading page when trying to open that. Yeah honestly as awesome as I find internet radio it's hard to pull something off with which I'd be satisfied. I have Azuracast, but just throwing in some ripped or maybe_not ripped music, semi good jingles and calling it a day seems not nice, but I don't have the capacity for anything else, nor would there be the interest by anyone.
@jonathan859 Internet Radio's ship sailed twenty years ago, for the average Joe really. As an assistant manager to a somewhat successful project, as in a peak of ten listeners, and now this defunct project of litterally zero listeners, I can definitely tell you that Internet Radio is a very niche, for select groups, market now. Its popular in the blind community for reasons, of course, but that's about as far as it goes. Even the majority of blindies are mainstreamed, and don't do Internet radio. Its Spotify, Deezer, Apple music, etc, etc. That and TuneIn, and they are even totally commercialized now, with a $1000/minimum per month to list.
Is that like how when I went to the Moog showroom in 2013, back when that was still actually interesting, I went in there fully expecting to play with synths all day, but spent most of my time running a sustainer guitar through the entire compliment of Moogerfooger effects? Also, I suck at playing guitar.
@BorrisInABox I'm surprised I never asked you if you've been to the factory or anything like that because of you being in NC. Mind you I grew up about half an hour away from AMS and only recently kfound it out.
@KaraLG84 Well, yes. I hear it's really boring now, because they don't actually make things here anymore. It is (was?) about a three hour drive from where I used to live/where I am now.
Whenever you find a doctor who'll treat that, put in a good word for me. Decent at playing bass here, there's often session gigs I should be taking, but I'm about 287% more motivated to piss about with code and occasionally go plinkity plonk on a synth.
Barrierefreiheit ist kein Extra und kein Bonus. Das ist die verdammte Unterkante von Qualität. Wer 2026 noch Webseiten oder Apps ohne vollständige Barrierefreiheit betreibt, ist fachlich ungeeignet. Punkt. Das ist kein „wir kommen später dazu“, das ist IT-Pfusch. Wer Mindeststandards nicht erfüllt, hat in professioneller Softwareentwicklung und Webdesign nichts verloren. #Barrierefreiheit #Accessibility #ITFail #Qualitätsstandard #WebDev
@BorrisInABox Hope you don't mind me poking you, but I'm pretty sure you know a bit about this. Do you know how I can LPC audio files using Windows? I had a look around and am not finding an alternative I can make work. Thanks for any help.
There's an online version of that utility someone made for Mac OS. I don't unfortunately know of a good tool to use locally other than SoX, which is not quite the same as the Texas Instruments implementation.
There is a Python utility, but I've never managed to make it work.
For the last nearly two weeks, almost all of my computing tasks have been done with my iPhone using NVDA Remote RC connected to one of two laptops back home in New York 500 miles away. It's just so cool to pull out a bluetooth keyboard and connect to a familiar environment from anywhere, but it's even cooler what you can still do without a keyboard if you don't have one, thanks to the custom gestures and presets for them. If you use Voiceover Announcements instead of AVSPEECH, you can even use a braille display, though it is nowhere as extensive as using a display direct with NVDA. So, if you absolutely rely on using a braille display, this is not the best way to work, but you are not completely without the ability to get an idea of what's going on.
I just did another small editing job remotely using Reaper on my laptop at home, and TeamTalk on a server running on a Raspberry Pi at home connected to Tailscale to connect the audio. Yep, it's an old TeamTalk server. I don't think there are rPi builds of modern ones anymore. Not the best way to do it, there is some lag, but considering I'm using Airpods Pro, which adds more lag, internet latency, etc. it worked pretty well, all things considered. Sonobus for iOS is, unfortunately, very jittery most of the time, probably because it hasn't been updated since 2023, and iOS has changed around it.
I have still not quite gotten fully comfortable with editing stuff on Reaper under Mac OS, at least with the same speed and efficiency, especially when it comes to envelope automation, and I have some plugins installed on the Windows laptop that I don't have on Mac OS that I need for this small weekly project, so it's nice to just be able to remote in and manipulate that existing environment. The fact that I can do that from my phone from anywhere, provided internet connectivity is good, and latency and jitter is reasonable, is just fantastic.
Right now, I have the second eSIM on my phone using AT&T, and amusingly enough, latency is almost the same to my stuff at home whether I'm using WiFi or AT&T. Jitter is worse on the mobile connection, but it's still perfectly usable, even for audio editing. I forgot that I had WiFi turned off for three days, didn't really notice.
And if you have a decent performing Android phone, you can get the audio latency down even further through the use of Soundwire. Provided that your interface has responsive-enough loopback that can be listened to through Wasapi. Soundwire unfortunately does not exist on iOS, since iOS doesn't expose the same platform-native low-latency network audio pasthrough needed for Soundwire.
I really have to give my laptop a static IP and play with this. I also found out I can somehow use it to remote control a Mac from Windows, which I very much want to play with.
@alexhall Yep, NVDA-RC for Mac can now be a target. And amusingly enough, using NVDA on Windows to control a Mac is a little more responsive than just using Voiceover by itself, because Apple.
I've got to agree with this. If I may respectfully add to it, it's not so much that we can do these things, we've been able to for a while, it's that #NVDARC makes it easy and smooth. It just works without workarounds. You don't have to think, just hit the icon and start typing. Note also that, when I say we've been able to do it in the past, I mean with the keyboard. Touch stuff is simply a new thing, and a very impressive thing with the speed and the smoothness. It really is astonishing in how well it works and what you can accomplish.
And another thing, with an android phone, you can connect a braille display directly using VirtualHere virtualhere.com All I can say about VO is that it works to some extent.
y'all guys, provide the best email service in the world, with extreme stability and zero-trust. If someone asking for more, he's greedy. You're appreciated, thank you!
-- tuta password manager -- Additional tuta free domains ("Tuta" has a bad, dorky sound to it in Swedish, keemail gets confused with gmail when you say it over the phone)
And improved handling of contacts/address book. I want tuta to be my "master" across Linux, Mac, eOS...
Android client: Problem: When an opened email is scrolled or "zoomed", hyperlinks are "too sensitive". (The links open in the default browser when I don't want them to...)
Solution: A setting that requires a URL to be tapped and held for 1 or 2 seconds before opening, or, A setting to cause a popup to "confirm before opening link".
Why? A malicious URL is currently too easily opened. Sometimes, users may wish to copy a URL and open it with a specific browser. Sometimes, users may not wish the sender to know they've opened the email.
Android client: Problem: When an opened email is scrolled or "zoomed", hyperlinks are "too sensitive". (The links open in the default browser when I don't want them to...)
Solution: A setting that requires a URL to be tapped and held for 1 or 2 seconds before opening, or, A setting to cause a popup to "confirm before opening link".
Why? A malicious URL is currently too easily opened. Sometimes, users may wish to copy a URL and open it with a specific browser. Sometimes, users may not wish the sender to know they've opened the email.
I've been playing around with your (paid) email and calendar services for over a year, but I cannot fully embrace Tuta due to its very limited customization options, especially for Calendar. I understand the lack of CalDAV synchronization support due to Tuta's end-to-end encryption model, however, personally I'll continue using a paid, private email provider whith full CalDAV support, so I could continue using dedicated Android Calendar apps such as aCalendar+, or Business Calendar.
Flatpak instead of AppImage only. I know there is an experimental Flatpak, but I would like a fully supported version that can also be found on your download page. It would be easier for new users and easier to push new users to both the Linux desktop and Tuta.
I'd like the calendar widget to follow am/pm vs military time. Other than that I feel like you all have added pretty much everything I wanted or are about to
@YvanDaSilva For you to never again advise people to use Brave browser, so that I can consider you a serious alternative to the ending of my Proton membership next October. 9 months. Show me the magic. I need proud and consequential intersectional antifascists as providers from now on.
To focus in your core product, mail. Handle attachments can be good with drive tools, edition of the files, etc... And stop the continous bait towards Proton, I changed my mail from proton to tuta, for the ia, crypto wallet stuff, and some vague political actions they had. Stand out if they do anything wrong, but in the meantime there has been too many post that goes against them without any purpose, and it is starting to be detrimental to your image. keep the good job:)
An improved email import feature is essential. The import feature of Proton mail, calendar, and contacts is the exemplar. Make the process as simple and slick as theirs. This will help organisations switch to Tuta, which is difficult at present.
It would also be very helpful if there were more quality replacements for Microsoft Teams and Zoom.
Sneaker auf der Bühne, aber ein Weltbild aus den 50ern?
In meinem neuen Artikel analysiere ich den „Hochglanz-Fundamentalismus“ moderner Freikirchen. Ich zeige auf, warum deren dualistische Weltsicht („Licht vs. Finsternis“) strukturell der Neuen Rechten in die Karten spielt.
I have a first laptop with HiDPI screen and trying to set GUI scaling in #Gnome.
I enabled experimntal fractional scaling, and set 166% scaling factor in Gnome settings.
For GDK apps there is no problem. But many other apps are not scaled, or are scaled partly - window header bar is small. Is there universal solution to this? Or how do you set scaling on all GUI toolkits like GTK2, Electron, QT, Java?
It's strange. The fractional scaling option should scale all windows. Of course, if the app cannot do it natively (e.g. because the framework it uses doesn't support it), then you're left with upscaling the window framebuffer.
yeah. For example the bitwarden app is scaled if I run it with env var GDK_SCALE=1.6 from terminal. So it can scale, but something is not giving the app proper instructins to scale? But I would expext that gnome gives the app all meeded env vars to scale properly. 🤷♂️
It isn't scaled at all, or it has an upscaled framebuffer (blurry content)? The former has to be some problem on your machine, it works for me. The latter happens. The app itself supports scaling, but it isn't compatible with the OS scaling.
but it made the electron app fractionaly scaled. 🤷♂️ It seems there is a mess between toolkits supporting other toolkits variables, or something like that. Well the prpblem is solved for me now. I enabled two experimental features, fractional scaling and xwayland scaling. Xwayland scaling is much less ready feature a caused strange scaling issues.
ahh, actually, try removing the 'xwayland-native-scaling' flag, and then restart the session. I noticed I have it disabled, and if I enable it, XWayland clients become miniature.
Yes, that's it. Thanks. Now xwayland apps are scaled, but blury. That's expected. With the expirimental xwayland-native-scaling some xwayland apps are scaled without bluring but with unscaled window header (vlc on the image), and majority is not scaled at all (bitwarden and others). 👍️
A masterpiece of progressive rock music from the famous LP Konvergencie, released in 1971 in the former Czechoslovakia; Opus, BratislavaLine-up / Musicians- ...
UK Prime Minister Starmer seeks support for international X ban
UK Prime Minister Keir Starmer is in talks with Canada and Australia in an effort to muster support for a potential international ban on social media X. It comes after the platform’s integrated AI assistant, Grok, was used to generate sexualised deepfakes of women and children.
Canada already ruled out a potential ban but Australia's prime minister was sterner in his rebuke of Grok, calling the AI-generated sexualised images "completely abhorrent"
Seriously. I have a bluetooth mouse that wouldn't enter pairing mode. It was connecting to *something*... maybe a smart switch or whatever. This cheap mouse has no way to enter pairing mode if it's connected to something.
So anyway, a microwave oven is a faraday cage. I turned the mouse on, tossed it in, shut the door and waited. It entered pairing mode quickly, and boom, connected to my tablet.
A+++ life hack by yours truly. Just don't start the nuker or magic smoke will come out 😆
New scrape: Instagram allegedly had 17M rows of largely public data scraped from an API and posted to a hacking forum this week. 6.2M rows also included an email address, and some rows a phone number. 100% were already in @haveibeenpwned. Read more: haveibeenpwned.com/Breach/Inst…
In January 2026, data allegedly scraped via an Instagram API was posted to a popular hacking forum. The dataset contained 17M rows of public Instagram information, including usernames, display names, account IDs, and in some cases, geolocation data.
Indonesia and Malaysia at the forefront of blocking Grok. I don't always agree with their policies on these things, but this one's worthy of respect. #TootSEA
"…muss ich bekennen, dass wir alle […] nicht ein Hundertstel, nicht ein Tausendstel dessen für möglich gehalten haben, was dann immer wenige Wochen später hereinbrechen sollte."
- Stefan Zweig, Die Welt von Gestern
Ein autobiografisches Werk Stefan Zweigs.
Das Buch entstand kurz vor Zweigs Tod in den letzten Jahren (von 1939 bis 1941) seines Exils und erschien postum 1942 in Stockholm.
German language. Stefan Zweig. DIE WELT VON GESTERN. THE WORLD OF YESTERDAY is the autobiography of Austrian writer Stefan Zweig. The most famous book on the...
hi i make goofy ahh tts vids for literally no reason lmfaoxd
Soar with your visions!! (EST. 2013, Early 2022)
Elecarlton is the home of unhinged, action-packed text-to-speech chaos and out-of-context madness, all starring the Epic Goddamn Microsoft …
@FreakyFwoof Nice. If I suddenly end up replying to this thread from my V Continuum account that means TC Social has an outage, only saying this because I'm noticing a few seconds lag when favoriting or replying to a post.
@FreakyFwoof In my MacOS days, I had your short number 91 as my startup sound. MacOS doesn't officially let you have a startup sound for the OS, so I pseudo hacked around it using only 3 lines of Python.
I'm not exactly sure why I'm doing this on a Sunday, and the hard work was done by others, but there you go; proposed fix for CVE-2026-0716. gitlab.gnome.org/GNOME/libsoup…
If the maximum incoming payload size is unset, then a malicious frame could cause an overflow when calculating the needed amount of data, leading to an out-of-bounds read...
Mal by som otázku/prosbu na mieste osadenstvo. Vždy som chcel vedieť programovať, ale nikdy mi to veľmi nešlo. Naposledy som to skúšal pred 13 rokmi, kedy som sa skúšal naučiť PHP (konkrétne Laravel).
Chcel by som to znova vyskúšať, a tak otázka znie, či by mi niekto vedel poradiť, ako na to, resp. nejaké tipy na návody a pod. Potreboval by som to vysvetľovať asi čo najprimitívnejšie ako to len pôjde, a ideálne aby to bolo v slovenčine/češtine. Taktiež aby to bolo zadarmo, keďže nemám peniaze na nejaké platené kurzy. Mohla by mi v učení pomôcť aj AI? Aj keď viem, že tu nie sú jej priaznivci, no ja osobne nie som jej zarytý kritik.
A aby som nezabudol, prioritne by som sa chcel asi naučiť to PHP (aby som vedel naprogramovať webstránky, nejaké redakčné systémy a pod.), prípadne ešte C++ (keďže by som chcel vedieť programovať Arduino).
Python je dobrá voľba z viacerých (netechnických) ohľadov, najmä dostupnosť učebného materiálu (dokumentácia, knihy, tutoriály, blogy) a veľkosť komunity (množstvo ľudí, ktorí vedia pomôcť).
Ako už bolo písané, je dobré mať nejaký cieľ (projekt), ktorý bude motiváciou učiť sa (áno, sú aj ľudia, ktorí sa učia teoretickú fyziku "len tak"...). Netreba sa báť skúšať, zlyhávať, pýtať sa a opravovať, dokým to nepôjde.
Ja "programujem" 2/3 svojho života a stále sa učím :)
Das letzte mal als ich versucht habe #NINA auf #GrapheneOS ohne Google Play Services zum Laufen zu bringen, habe ich es leider nicht hinbekommen (ist aber schon ein paar Jahre her). Wenn jemand passend zum #did #dut nach Alternativen sucht: #fosswarn
Mit FOSSwarn bekommt ihr offizielle Warnmeldungen für eure Orte direkt aufs Handy. Push-Benachrichtigungen funktionieren ebenfalls zuverlässig über UnifiedPush, ich nutze dazu die App #Sunup
(if you don't see it: there's a tiiiiiiiiny difference… 60nm tiny to be exact… and these 2 interfaces are in a LAG and actually work, the other side is even more insidious since it just says 1310nm but the SFPs *are* different)
Dutch AON FttH specs are saying just TX 1310nm, RX 1490/1550 nm. I bought a 1550nm one as they are much more common and got no link. So I bought a 1490nm one and the link came up immediately.
Then I bought fibre cleaning pens, got extra 10 dB by cleaning the connectors and now even the 1550nm one gets the link. So probably if you have enough optical power, all SFPs are wideband 😁
Walmart partners with Google to offer AI-enhanced shopping, allowing users to purchase Walmart and Sam's Club items directly on Gemini (Jaewon Kang/Bloomberg)
A look at the history and rise of Markdown, which has been adopted across the tech industry in the decades since its quiet launch in 2004 by John Gruber (Anil Dash)
dělal na tom před pár lety kolega. Ve skutečnosti první najede obrazovka UEFI a pak se zobrazí bootovací obrazovka systému, díky kolegovi si to dokáže vzít to logo z UEFI a zdánlivě to vypadá, že to je ta samá obrazovka, jen tam přibude logo systému.
“Code is not an infinitely reproducible machine that requires no labor inputs to operate. Rather, it is a brittle machine that requires increasingly heroic measures to keep it in good working order, and which eventually does "wear out" (in the sense of needing a top-to-bottom refactoring).”
The myth that "old code doesn't rust" persists because it justifies moving fast, breaking things, and leaving it for someone else to clean up the mess.
@katzenmann@c3d2.social That's exactly my point - the "whole Linux community doing Wayland" was never true to begin with. It was Red Hat pushing it like crazy and KDE unfortunately following suit.
small reminder that this user is the one who wrote "Wayland breaks everything" and never seriously updated it to deceive people into thinking Wayland is broken etc.
This repository contains resources concerning unofficial Linux distribution aimed at visually impaired users. This distribution is called Vojtux. The name is based on first name of the main contributor; Vojtech. github.com/vojtapolasek/vojtux
"Desde los años 90, existen tecnologías que permiten que una persona con discapacidad visual pueda acceder a un ordenador y usarlo con total normalidad.
I mean—I mean—I MEAN—that just— I'm sittin' here on GitHub—I mean I'm sittin' here, looking at the CLAUDE.md in your GitHub repo, because you'll write decent docs for the computer to read that I couldn't get you to write for me or junior devs for the last quarter century
Timothy Wynn
in reply to Tamas G • • •Jayson Smith
in reply to Tamas G • • •Tamas G
in reply to Jayson Smith • • •FL (or the Speech Synth plugin/module) calls ZXSpeakInit in speaker.dll.
Then, speaker.dll loads/uses svctl (“control” layer), and svctl loads/uses sveng (the actual voice engine). (This is implied by the dependency on SVCTL32.dll and the “3 file” requirement.)
speaker.dll then asks SoftVoice to synthesize the text, likely into a PCM buffer.
Finally, speaker.dll (maybe via MSACM32.dll) converts/normalizes the audio into a format FL expects.
So in theory inspecting details within speaker.dll can give all the clues too, but reading disasmed dll is so not een in my pervue, lining up function names to areas of the DLL, loading it in debuggers to check what it dos in memory, ETC...
True about DecTalk though. It's why I was surprised when Blazie used it in BTSpeak, though perhaps there's some other licensing / conversations we don't know of there so I won't assume anything. Still, it's either that or clear they saw low-risk at distributing it more commercially within BTSpeak units. Very interesting. And yet, Apple themselves probably will never touch it, despite it being so in the open in sources even if unofficially, they just know they can't touch it with the historical significance and potential licensing rights for them.
Jayson Smith
in reply to Tamas G • • •Tamas G
in reply to Jayson Smith • • •Jayson Smith
in reply to Tamas G • • •Kara Goldfinch
in reply to Tamas G • • •