For those who wanted a SAPI5 version of NVSpeechPlayer, it is ready. You can install the file from: eurpod.com/synths/NVSpeechSapi… This will install it as both a 32-bit and 64-bit executable file. It will also create a settings program that lets you configure which languages to uncheck from your voices list, this should be in the start menu. The engine should work in both 32-bit and 64-bit programs.
Hmmm. Any idea what this means in my NVDA log? I go to sapi 5 in the set synth and get cannot start the sapi 5 synthesizer. This is the log: Traceback (most recent call last): File "synthDriverHandler.pyc", line 514, in setSynth File "synthDriverHandler.pyc", line 481, in getSynthInstance File "synthDriverHandler.pyc", line 370, in initSettings File "synthDriverHandler.pyc", line 385, in loadSettings File "synthDriverHandler.pyc", line 429, in changeVoice File "synthDrivers\sapi5.pyc", line 767, in _set_voice File "synthDrivers\sapi5.pyc", line 642, in _getVoiceTokens _ctypes.COMError: (-2147200967, None, (None, None, None, 0, None)) INFO - synthDriverHandler.setSynth (10:32:58.615) - MainThread (4976): Falling back to previous synthDriver ibmeci I know for a fact I have sapi voices and I've repared the registry countless times to no avail.
@TomGrant91 Yeah — that looks like a SAPI/COM registration (or bitness) problem, not an NVDA “settings” bug. _ctypes.COMError: (-2147200967, …) is NVDA failing while enumerating SAPI voice tokens (_getVoiceTokens). When someone says “I know I have SAPI voices” but NVDA can’t even list them, the usual culprits are: • Wrong bitness: 32-bit NVDA can only see 32-bit SAPI voices. • Broken SAPI token registry hive (or missing permissions). • Corrupted voice token (one bad token can make enumeration blow up). • SpeechUX works but NVDA doesn’t: often means only 64-bit tokens exist. Try to see if the voices play.. run: C:\windows\syswow64\speech\SpeechUX\sapi.cpl (this is the 32-bit speech panel.) Also you can try running regsvr32 on the dll to see if that fixes it - if it does, I need to check the install script and make sure it does this. for that run: C:\windows\syswow64\regsvr32.exe "C:\Program Files\NVSpeechSapi\x86\NVSpeechSapi.dll" to register the x86, and C:\windows\system32\regsvr32.exe "C:\Program Files\NVSpeechSapi\x64\NVSpeechSapi.dll"
I have literally got back from my activities. I will try these in my own time. Thanks for your help though and I'll let you know. I've had this issue for a while, not just the speechplayer.
@TomGrant91 I remember Jake having a file on his site for Sapi Registry Fix or something similar. Like Sapi Fix. That might actually help you here as it can help re-enumerate the engines you've got installed and recreate the right voice token data in the registry for each.
"Mit Work-Life-Balance und Vier-Tage-Woche lässt sich der Wohlstand unseres Landes, den wir heute haben, in Zukunft nicht erhalten - und deswegen müssen wir mehr arbeiten", sagt Friedrich Merz.
@Bri @BorrisInABox This is what stereo coiled bop it arcade sounds like. It is, as expected mostly mono content, but a couple of parts are stereo. This is slightly panned right from the daw as it was recorded slightly more left heavy when I recorded this time. I got a score of 68, which is my new personal best. That's great, but also at the same time it's annoying since they set the bonus score on that machine to 75, so if I had just been able to get that few more points, which would have included an even faster speed, I could have gottent eh victory sound. I also figured out when going to record fail states that on level one, they give you up to 2 retry attempts, so that you hopefully don't end up wasting a credit on a poorly played game.
the fact that it give you retry attempts on level means tha i was able to get a snapshot of the music on level one which was mostly clea accept for the voice command at the start, but mvsep took care of removing it.
I am curious, so today, lets do some Mastodon research on typing verbosity on your #ScreenReader, when typing on a physical keyboard. What is your Typing echo set to? #accessibility #a11y
No, I am not going to do a search for the great Gatsby fanfiction, but, for some reason, that just came to mine. I mean, it is public domain, has been for a while, so it most likely exists in bulk. There is probably a huge, closet club of the great Gatsby fanfiction writers out there Just having a great old time fulfilling their Myrtle fantasies, but I will never know. Why am I even thinking about this?
Also, fun fact: that book is why I failed English in the 11th grade. I just didn't care about it, didn't do any of the assigned stuff on it, subsequently failed the class. YAY!
Things you probably shouldn't hear in a grayhound station, but actually do. A female announcer making a bus announcement. Bus 2501 departing for various stops in Tennessee, Now boarding at gate d3 d as in Dick! The hole dam station just went so quiet you could have heard a pin drop.
Oh good, we're alive again. It would really be nice if Postgress didn't start randomly doing this thing where it decides to use so much ram that the OOM killer owns it, that it never did before, but I guess I'm going to have to spend two hours Googling to fix this bullshit.
Anyone happen to know how to get SoX on a Synology DS1815+? I have a bunch of things I want to automate on a schedule that needs it specifically. I could mount the filesystem and do all that on another host, but really don't want to do it that way. I might be able to do what I want with ffmpeg, which is installed, but would rather use SoX if I can.
@ner Don't know. I can try copying a 64-bit binary from any number of things, but I don't know if it's self-contained/static linked/whatever without trying it first, I guess.
Apparently, my copy of build-essentials got nuked with an update. Synology likes to do that. Hold on a bit more. Now I see Andre's deleted my login ... again. He likes to go an delete the home folder on that thing. Hoping he hasn't killed the blind hams stuff, another great reason why we need our own damn server
You just need the static sox. Chat gpt recommends pulling it from a throwaway docker image of Alpine, that'd work, but as long as you have a static copy and know the commands, it'll work. If you need a man page, then ... But a static sox will include the libraries you need. Let me know and I can pull it, no need for you to build.
Just be glad you don't have my old 1104J sunology that was Marvel and you had to build things for Marvel, and not just any Marvel, a weird and stupid version of the proc
@ner Stupid and dumb... All I need it to do, really, is process a bunch of raw G711U files and put wave headers on them with the appropriate chunks and crap. It's super uncomplicated. Like, I could probably actually do it with a bash script that writes out a header for each file based on the filesize. But yeah. I have about 50000 of these things.
Let me know if it works, it does on my synology, I tried it. Though i have a newer 1821+, we have similar processors. I ended up building this in ubuntu
@ner This is what I got when I tried to run ./sox --version from the directory it lives in. ./sox: error while loading shared libraries: libltdl.so.7: cannot open shared object file: No such file or directory
So I'm having to build a crap ton of other stuff in an older system. That was after copying newer libraries which synology doesn't like. I'll get it working on my system, hopefully it won't burp and die again
@ner I've been seeing the thread, and unfortunately have zero experience putting software on Synology NAS that didn't come with it. However, doesn't it support optware or entware? So far as I'm aware, they are package management and build environments? But it also sounds like you two will have it resolved soon. But maybe these other things are workable ideas.
This is a bundle with libraries, uncompress it to a directory somewhere (making this totally static was really difficult/not recommended. When you have it in the folder you wish, test sox --version which will come up with 14.4.2, and then simlink to a user bin or somewhere. It'll get the libraries which I custom built in the uncompressed sox-bundle. Then you can do all the scheduling junk / add simlink to a path folder and bingo.
@ner What I really need to get working on the NAS is stuff like screen and ed. Right now I'm having a local pi be a screen and such, and it works, but heh. Oh and scheduled tasks that don't suck.
Too few people really think about the continuum between gravy, stew and soup. To really ponder, when you get right down to it in life: is the liquid or the solid bits more important in this meal?
I was apparently arrested at age 20 for committing arson, killing 8 out of 10 people in a building. It was ruled self-defense, as I was escaping a horrible situation.
I was apparently traumatized by this whole deal, and had to live in a group home, where I met a girl who also murdered in self-defense.
Things were going well.
Just as we were really starting to get to know each other, I woke up, and the dog was frantically barking about something, so I guess I'll never know how things turned out.
Grammy® award-winning Chanticleer releases its first music video - Indie Pop hit "Cells Planets". Exploring new music has been a defining characteristic of ...
@Bri You caaaan! Also, kick-ass for bringing this to client as quickly as you did. I was thinking it wasn't gonna be added for at least a few months, but yo. Killer shit here.
Yeah. And the cool thing is, the way I implemented it, it shouldn't break as long as YT-dlp works. You can update yt-dlp separately from the actual app.
actually, though, you know what, it looks like VLC by itself does support youtube playback, but if that stops working yt-dlp is implemented as a fallback if you have it
Physical displays or virtual windows? For physical displays, pressing Windows+shift+arrows moves a window between monitors and reports it, but I'm not sure how to query which display a window is on without moving it.
For virtual windows, If you press Windows+tab it opens the task view window with the current app active. Press tab once to the list of desktops. The focus starts on the first desktop. Press right arrow to move through and the one which doesn't say "not selected" is active
When I did win+shift+left and right, or win+alt/left and right, not entirely sure what the difference is, both NVDA and Narrator report the window is now on "display 1 2". it seems to report this for both displays so I'm not sure how that is meant to be helpful
Maybe an oddity in your version of Windows (since both Narraotr and NVDA report it)? I just tried again here and it definitely reads "WINDOW TITLE maximised on display 1" or "...on display 2". I'm using Windows 11 25h2 build 26200.7623 with the screen set to "extend". Actually, could you have your window NOT maximised, but windowed and showing partly on both screens?
Ok, well, it's 4 in the afternoon, almost evening, and how far is NVSpeech Sapi? It speaks a single sentence and then crashes. There. That's it, one peep does come out of it now. :D
I need to lenghten a bit the i before ã in words like rebelião, religião, avião, união. It is too short. Can I do it by modifying a setting, or only you can do that?
@clv1 hm, this is where the lengthenedScale setting can help you, the default is very short but you can change it for portuguese if it is helpful for that language to be something like 1.14. Try that first too. You can add this rule to the PT language file, rules don't require a restart of NVDA, just a restart of the engine (so switch synthesizers, switch back, rule applies.) So in Pt.YAML inside packs\lang for the add-on: # --- Portuguese "-ião" glide length --- # In words like "avião" / "religião" eSpeak emits ...iˈɐ̃ʊ̃. # The /i/ here behaves like a glide into the nasal diphthong, but it can sound too short # in our timing model. Add a length mark so the engine gives it a little more time. - from: "iˈɐ̃ʊ̃" to: "iːˈɐ̃ʊ̃"
@Tamasg Thanks a lot. It appears the best compromise is around 2.44 for scale lenght. Also, I tried adding your mapping suggestion in the phoneme editor, but it wasn't accepted, perhaps because there is more than one key for the To field. And speaking of nasal diphthongs, I'm not quite pleased with ã and õ. Tomorrow I'm gonna try cloning and modifying them.
Wait, are you calling yt-dlp at all? Or can VLC do that? Or does it rely on system yt-dlp? I've got one of those, uv tool from the commits, which is basically nightly.
I recommend shutil.which. Works on Windows, finds yt-dlp in your path. That's sort of what foo youtube does for the system option. Mine's in ~/.local/bin as part of uv tool.
Guess you need a link to the zip in the docs, and see about a mac build of said. But since I have VLC and yt-dlp both installed, it should just magically work?
Well, except for when youtube videos require PO tokens or authentication or whatever the hell, but there's not a whole lot you can do about that. Long as you don't pass it some option that says please ignore all configurations if someone has, say, a PO token provider configured that should work too.
@x0 I think they have flags now that can take cookies from (some) web browsers. It's a very dangerous thing to do (people get their Google accounts banned for this), but necessary in some places due to the age restriction laws.
if you are talking in your program, I just have one question. I thought I saw somewehre that it uses vlc to do it. if I use vlc to stream to one of the kids tv's, wouldn't that mess up what is streaming to there tv?
@JamminJerry appreciate it. If you feel like it while I'm gone, try to go into global options, advanced, and then hit download yt-dlp. Once it downloads, try to play it again. If that still doesn't work, try to hit the download VLC button in that same advanced dialog. Once *that8 finishes, try a second time. Let me know your results.
@J3317 actually I try to keep my laptop extremely simple, and as I don't have the same keyboard layout in his program, and his doesn't have telegram inj it. I just keep it with what I know. I don't want to be half awake, and try using what I think is telegram in tweesecake, and accidently send out something on mastodon that would be rather imbarrasing. lol.
I am almost afraid to find out. I mean what if I find cam posting a cop video he just thinks I should watch, and at the time Kalvin is watching paw patrol on his tv. I would hate to have it start streaming that cop video to kalvins tv. ROFL!
Asked Copilot to build a web app and I am not surprised to see it generate tons of DIVs instead of semantic HTML, but hard-coding the type of element into the JS variable is taking the mickey…
Hey Everyone. Tonight at 6 PM Eastern, Sonic Synergy goes live yet again and offers you a heapin' helpin' of crazy conversation. Not sure what we'll discuss this time, but isn't that part of the adventure? And since our show is powered by your calls and texts to (833) 452-4452, you can help drive the ship.
Gosh, with all these tools we should already be so fsafe beyond safety's own limits. Why do we then need ICE and all these oso-called law enforcement agencies? I mean just put a guy in a cowboy hat and boots, show them how to use Tangles, and the rest of us can go party. What's wrong with this formula? @TexasObservertexasobserver.social/@TexasObs…
Thanks to a software program called "Tangles," #police—sheriffs, Texas state cops, and even #ICE—can follow the location of your smartphone without a warrant.
One sheriff who leads an anti-smuggling task force says the software helps “develop leads to eventually obtain probable cause.” Civil liberties experts say its use violates constitutional rights.
Amazing news! NVSP Phoneme Editor V4 is here, and it brings much needed improvements. For starters, we directly consume libespeak.dll, thanks to learnings from our SAPI5 wrapper. No more running Espeak in another process this way, but still side-loading the DLL into the process. The .exe is included so as to keep the distribution of Espeak intact. This also allows us to get correct IPA mappings just like the NVDA driver gets, so letters should not be dropped during speech. Check it out! Wow! Also, the spin controls can now use page up and page-down for movement by increments of 50. This should allow wider mapping but finer if you arrow. Wait, spin control? That's right, hitting "edit" on a phoneme will let you increment it using the arrow keys, or type in the value. Why not be flexible! We solved an issue with loading: no matter where you unzip this, it will load both the DLL and Espeak NG folders, locating Espeak-NG.exe in it. There's still a bit of a bug here because we don't check for libespeak.dll but the executable, but this is a minor one for the next version to fix. If you point it at a dll, it will also load the packs if it exists in the same folder, which also means less hitting "open" and "set" everywhere. Enjoy! eurpod.com/synths/NVSPPhonemeE… And yes, I tested this with at least 4 KB of text thrown at it at once in the text-box, it didn't crash. I'm sure there's a limit, but still, impressive it only took like 7 seconds to generate the speech.
I just tried version 4 of the phoneme editor, and I'm trying to load the packs from the folder C:\Users\B\AppData\Roaming\nvda\addons\nvSpeechPlayer\synthDrivers\nvSpeechPlayer and when I press the select folder button, the application crashes. I was hoping to use it to modify a few phonemes in US English that sound off to me but I'm getting this crash. Thank you very much for your time and I look forward to any input you may have. Thanks!
Thanks to a software program called "Tangles," #police—sheriffs, Texas state cops, and even #ICE—can follow the location of your smartphone without a warrant.
One sheriff who leads an anti-smuggling task force says the software helps “develop leads to eventually obtain probable cause.” Civil liberties experts say its use violates constitutional rights.
I love YouTube, and by that I mean watching it; I would never upload there, especially now, with all of their silly censorship of things that don't even need censoring.
Attached: 1 image
The jQuery team has released version 4.0, introducing breaking changes, modern browser support, and long-planned cleanups.
https://linuxiac.
When you spend a couple of hours trying to troubleshoot program paths, because you heard, uv as, of with your screen reader and kept typing it as, of, instead of, uv. So annoying! That's why I like braille access!
One thing that's tough for people who just join the Fediverse, like most of the folks on #theforkiverse , is that there are self-appointed experts who reply to their public posts saying that "the Fediverse" doesn't like that kind of content.
This is untrue. They don't speak for all of us. Nobody does. You can't spell "Fediverse" without "diverse"; for good or ill, there are different viewpoints on almost any topic found here.
Had to resort to Mona on iOS to edit a private DM. I couldn't find the option in fast SM or the web , but maybe I wasn't looking in the right place, but I know how to do it for regular posts soooo. Maybe it was user error though.
I will remember that in future, thank you. Didn’t help that I was dealing with Google. Logging me out of my Gmail for security reasons. At least I can still login on my known devices: Hopefully that will get resolved soon.
I'm helping an elderly person with a patient portal, and wow, there are about 4 different problems ranging from unclear instructions to pages that don't work well when the phone is zoomed in enough for an older person to actually read the text.
Plus, the iPhone keeps popping up unhelpful gunk.
This person isn't particularly afraid of technology either, but they literally can't do it.
I now think every web developer should be forced to walk through their processes with an 85-year-old.
*Round of applause, some cheerful whistles and yelling can be heard* Welcome to the club of people needing accessibility! We struggle with such things every single day. I mean, "we" as in "blind people" even more, but still 😊 On a more serious point, try reporting the issues to the dev. As you are in the US, your strong words to mention are ADA (Americans with Disabilities Act), Section 508 and WCAG.
Accessibility is not an excuse. For some backstory: I was at lunch with my parents when they told me I could do classes online but would most likely have to pay for it to which I said, "I would have to do some checking to make sure the platform was accessible." My dad then said one of the worst things anyone could ever say, family or not. "Oh, there's always an excuse." Accessibility. is. not. an excuse. It might seem like it to someone who might not have to think about something like that, but this attitude is disgusting. Accessibility comes in many forms, and having to think about whether something is going to be accessible or not is so goddamn tiring and I don't like that thought being the first thing I say, but also, that's a reality I have to accept. Again I'll say, accessibility, no matter what form you're thinking about is no excuse. After this interaction, I'm left with -1 spoons and a desire to do nothing else for the day.
Sometimes you just have to jump in, yes not everything might be the best decision to make, but honestly getting a Raspberry Pi and not waiting for the perfect moment to start selfhosting when, I had better hardware, or more money, or better internet, or lived in my own home etc, was one of my best decisions 2025. I love beeing able to spin up seriously useful docker containers in seconds, and only today, I added Beszel and linkding to the list of useful tools amongst gitea, audiobookshelf, joplin. Not only does it safe money (to host a joplin server instead of paying for joplin cloud, for example) it's fun, it's satesfying, it takes almost no resources. Just had to say it again, this is good.
Honestly I'd do this more, if I had a way of backing up all this stuff. Like right now I kinda trust all the big companies to at least cover that if something happens LOL.
@johann Fair, ngl I haven't setup backups yet but it's a high priority for sure. In the end it's only running for me so that's the good part. But I'll probably just use Kopia to backup to a Hetzner storage box, could be any other external storage place though. It'll encrypt the stuff so I don't have to worry about any company sniffing through my data.
Yeh seams like the best solution for this case, use something like Kopia or Restic to backup to remote storage. I just need to set it up properly, just have to get over the lazyness with actually looking at how to use these things.
Following the example of María Corina Machado, who gave her Nobel Peace Prize, Arnold J. Rimmer also gave his Bronze Swimming Certificate to Donald J. Trump.
Anybody have ideas about where to host a podcast (low traffic)?
One thousand years ago, I started a thing on a podcast service that got acquired by Spotify. The podcast is served from Spotify today. I’d like to change that.
I did figure out a story for hosting the podcast feed directly from my Ghost dot org site, but it’s a bit too fiddly for me to want to take that on right now, especially with a back catalog of episodes I’d want to keep online (ideally without migrating each item manually).
.
Tom Grant
in reply to Tamas G • • •Traceback (most recent call last):
File "synthDriverHandler.pyc", line 514, in setSynth
File "synthDriverHandler.pyc", line 481, in getSynthInstance
File "synthDriverHandler.pyc", line 370, in initSettings
File "synthDriverHandler.pyc", line 385, in loadSettings
File "synthDriverHandler.pyc", line 429, in changeVoice
File "synthDrivers\sapi5.pyc", line 767, in _set_voice
File "synthDrivers\sapi5.pyc", line 642, in _getVoiceTokens
_ctypes.COMError: (-2147200967, None, (None, None, None, 0, None))
INFO - synthDriverHandler.setSynth (10:32:58.615) - MainThread (4976):
Falling back to previous synthDriver ibmeci
I know for a fact I have sapi voices and I've repared the registry countless times to no avail.
Tamas G
in reply to Tom Grant • • •_ctypes.COMError: (-2147200967, …) is NVDA failing while enumerating SAPI voice tokens (_getVoiceTokens). When someone says “I know I have SAPI voices” but NVDA can’t even list them, the usual culprits are:
•
Wrong bitness: 32-bit NVDA can only see 32-bit SAPI voices.
•
Broken SAPI token registry hive (or missing permissions).
•
Corrupted voice token (one bad token can make enumeration blow up).
•
SpeechUX works but NVDA doesn’t: often means only 64-bit tokens exist.
Try to see if the voices play.. run: C:\windows\syswow64\speech\SpeechUX\sapi.cpl (this is the 32-bit speech panel.)
Also you can try running regsvr32 on the dll to see if that fixes it - if it does, I need to check the install script and make sure it does this. for that run:
C:\windows\syswow64\regsvr32.exe "C:\Program Files\NVSpeechSapi\x86\NVSpeechSapi.dll"
to register the x86, and
C:\windows\system32\regsvr32.exe "C:\Program Files\NVSpeechSapi\x64\NVSpeechSapi.dll"
Tom Grant
in reply to Tamas G • • •Tamas G
in reply to Tom Grant • • •Tom Grant
in reply to Tamas G • • •