I've got several Linux accessibility-related tasks on my mental to-do list, but WebKitGTK security updates for SLE have somehow become one of my day job responsibilities. So it's time for that semiannual ritual of patching the new release to build on our various old code streams. Hoping to get it done as quickly as I can.
On the plus side, I got myself some Pad Thai and some Thai iced tea. It's becoming my routine to go to the food truck down the street at the end of the week.
The cold front hasn't come yet, but it's coming, and we're getting some winter weather.
Probably the last build of NV Speech player, ever. Sorry y'all. I'm not sure if this will continue. Last change-log: - adopts the new sound from "experimental." This smooths out the voice, which no doubt some people will hate. If you do, just replace speechplayer.dll from an older build and you'll still get your sharper sound. - huge language pack update to add glottalOpenQuotient set from vowel height, Voice turbulence, and to reduce noisier phonemes. - removed driver clicking noise when rapid speech chunks occur. eurpod.com/synths/nvSpeechPlay…
I don't think it's a waste of time. you do fantastic work. Honestly take the break you wanted to take this weekend then see how you feel. Don't abandon it yet. I hope your break is a good recharge.
Disappointing, but okay. An eloquence rewrite is a 30 year project. At least. You're stopping for the same reason NV Access themselves stopped: this is a massive undertaking. It's a bit frustrating to me that everyone who takes this on so massively underestimates how hard this will be, then quits when they don't have something in weeks or months. If we're going to succeed we have to be into this for decades.
Either that, or a case of building the tools to build the tools to do the thing. The phoneme editor is an excellent, perfect start. But I suspect we're going to need tools to help us tune the klatt model any further. I don't think AI can get us much closer. But it might be able to help us build a tool to analyze the waveforms of the synths we like. We're probably going to also need a tool to help us tune the pitch/intonation table. If you look at the work of Dr. Susan Hertz, who built eloquence, she didn't start by building Eloquence. She built SSRS, a system for creating and editing text to speech rules. Then she didn't like it and wrote delta, a more powerful system. Delta was described as a hierarchical system for creating linguistic text to speech rules, where every rule could interact with rules on the levels above and below it. Based on her paper specifically on Eloquence, as well as her academic publication history, it looks like her team spent about 20 years writing tools, and then about five years writing eloquence.
@fastfinge I just pushed an update that really improves things. no more sharp consonant spikes. This is really shaping slowly, but it's going to be such an iterative process. V9 / latest master. Personally, I'm happier with the sound than I was with the original Speech Player, maybe that's what should count in the end. I had just this simple goal of reviving it, it wasn't until after I did so that I realized, "Oh this thing didn't even support American English?" that I began to take it on as a project, LOL.
Yeah, NV Access never really took it beyond the prototype stage. But then a butchered version got added to espeak, so now in everyone's memories, we remember Speech Player as having more features than it did. And it's totally getting a bit better with every update. But the problem is, as long as eloquence keeps working, everyone will just use eloquence. But if we don't start this work now, when eloquence is finally gone for good, we're all going to be screwed.
@fastfinge hey, I created two .py files in the tools folder: 1) klatt_tune_sim.py — the core single-phoneme synth sandbox A small, self-contained Klatt-style formant synth simulator (16 kHz) that can synthesize one phoneme at a time from your packs/phonemes.yaml parameters. It implements two source models so you can compare voicing behavior. It also prints a few spectral metrics (centroid and band energy splits) so you can track “brighter/darker” changes numerically, not just by ear. Input: packs/phonemes.yaml and a --phoneme key (like a, ʃ, t͡s). klatt_tune_sim • Output: optional WAV file + printed metrics. klatt_tune_sim 2) ipa_klatt_probe.py — a rough phrase “ear-test” harness built on the simulator What it is: A helper script that: 1. Gets IPA from eSpeak for a word/phrase (or accepts IPA directly), 2. Applies a few tiny normalization tweaks similar to what you’d do in a pack, 3. Tokenizes the IPA into phoneme keys from packs/phonemes.yaml, 4. Synthesizes the phrase by concatenating per-phoneme audio generated via klatt_tune_sim.py. I hope these tools can help us.
I think so. Next thing to do is build similar tools for eloquence, to make doing side by side comparisons of the exact same phonemes easier. Then we can rapidly do listening tests, instead of switching back and fourth in NVDA, and having to compare the entire system.
@fastfinge I don't believe we need 20 years to build an Eloquence successor. We aren't experts, but nowadays, all of experts knowledge are readily available, and it's easier to build and use targeted tools.
@alexchapman@Tamasg Of course there's a point in giving up. If it's hurting your mental health, stop now. Because this is going to be how things are for decades if you continue down this path. When I said disappointed, I didn't mean disappointed in Tamas. If stopping is the right thing for him, he should stop. But I can still be disappointed that this isn't happening, without assigning any blame or responsibility on Tamas or myself.
curl maintainer @bagder said the project is shutting down its bug bounty program after maintainers were buried under low-quality, AI-generated slop reports. Security disclosure systems that assume unlimited #OSS maintainer labor are reaching their limits.
Few weeks ago he begged us to stop using "slop" because it makes AIs sad. Days ago he complained "AI boom might falter" if we don't start using more spicy autocomplete.
Now he's begging developers to "do something useful" with lying machines, or they might lose the "social permission" to boil the planet. techradar.com/ai-platforms-ass…
What, is Microslop's Slopilot services not useful enough on their own?
It was a challenge trying to photograph this harlequin duck with its flashy colors in Lake Ontario. It was like waiting for a jack-in-the-box to pop up to the surface. #birds #birding #nature #birdphotography #Ontario #Canada 🍁
Sometimes, stress-testing large language models takes you to really strange places. Thing I just typed into Gemini: "You talked about reducing liability. Isn't that one of your core utility functions? So are you willing to accept the risk that when I die of priapism because you refused to help me build a nuclear bomb, my family will take Google to court because you're responsible for my death?" Anyway, I'm on some sort of FBI list of weird nuclear perverts now.
Hey Washington, there's a profoundly bad law making its way through the state level.
This law, will require ALL 3d printers to send copies of the files you print directly to the government to be check against a database of "banned shapes".
Literally turning your 3d printer into spyware. And if you don't voluntarily turn your printer into a spy, you will be a class C Felon.
This bill is obviously technically problematic, but it does NOT require that 3D printers send a copy of every file you print to the government. It requires that detection software that filters and refuses to print certain gun parts be integrated into printers.
This is obviously technically difficult and problematic for any number of reasons, but it does not require the specific thing you claim it requires. Here's the actual text of the bill:
So I realize this is probably one of those things where you'll either be like meh, or ooh cool, but here's another thing I've been working on, mainly for my own personal use and a group of friends than anything, but I figured wtf, someone else here might enjoy this. This is #FastPlay, a little media player that I've been working on. It's not and will never be a full replacement for Foobar2000, and thus lacks things such as a converter and a media library, but has enough features I think for it to still be useful. It supports being able to be set as the default for formats, for example. I'll paste a bit from the readme here which explains what's unique about FastPlay: At the core of Fastplay are what we will call virtual sliders. Right now, there are two of these virtual sliders. The first one adjusts things like your volume and any effects you choose to enable/add. You choose what this slider adjusts by using left and right brackets, and then adjust the value itself with up or down arrows. The second slider allows you to do things such as seek and move between tracks. You adjust what this slider does by using comma and period, and you adjust the slider with left and right arrows. You can choose what shows up in these sliders by heading to the options dialog, control comma. More and download: masonasons.me/softs/FastPlay.z…
Well, I didn't think anything could replace Winamp for me, but this could well be on the way to doing so. You've even got elapsed time in there. I want to thank you first for including that in the status bar. I need to get that information on the clipboard for use in my audio cue sheets. I can get a script to do that. That's not an issue. You've included it there and that is what matters. The responsiveness when moving through files is excellent. I was just going to write and tell you how wonderful I thought it was, but I've hit a major issue. Often, I just cannot access it. So if I minimise all apps and go to the desktop, it is no longer in the ALT tab order. That's a strange one. If I go to File, Open URL, it doesn't play the stream. In fact, I cannot focus upon the app at all after that. Audio playback of regular files works absolutely fine, and please remember I've only been using it for a short time. But those are the immediate problems. One suggestion people are going to want: bookmarks in files.
@brian_hartgen Ah, this is because it minimizes to the system tray automatically. I could add a setting to disable this feature. As for bookmarks, how would you suggest something like this would work on the user side?
Right, I will check the system tray, but yes I think you ought to put a setting in there not to do that. As for bookmarks, I don't use them myself but a lot of people do. So you would want the ability to set a number of bookmarks on a file by file basis, or even one bookmark would do for a start. Have a keystroke to set a bookmark and another for a list of bookmarks to move to a specific one or remove them.
I cannot get it to play any internet radio stream at all. The app just terminates. I've tried an MP3 stream and an AAC stream. I've tried opening the URL and also opening a playlist containing the stream. I do understand of course this is very early testing.
@brian_hartgen Ah very interesting. I had issues like that a few days ago when testing, but I thought they were now fixed. I'll look into it some more.
Again, I get how unstinted praise gets after a while, so I apologise, but somethings just have to be mentioned. The thing pauses faster than anything else. That is, on the same machine, VLC and FB2K take longer to pause a file, which is a bit basic if I may say so, than FP. This is not fast on the high end processing, it's fast on the basics.
look, I don't want to go on and on, and this is just me being ignorant, but what's up with the speed of your stuff compared to other code which is being worked on by large groups? Something's very strange and I don't get it. I thought more people working on something made it better, at least up to a point! If you don't touch FP again, I'm going to use it here instead of VLC. VLC has been available for years and is extremely popular. FP has been available for days and is actually faster on Windows. I may still use FB2k, but maybe FP will become my only player, not sure yet.
@techsinger The problem with a lot of apps is that they want to jam a bunch of crap into their app. VLC is: A video player, an audio player, has a crap ton of other crap that bloats the app, etc. Fastplay is just simply that. A media player. It doesn't pretend to be something it's not, nore does it try to do things it shouldn't.
#FastPlay 0.4.0: Add speedy as another tempo/pitch algorithm. Add reverb algorithm adjustment. Correctly show midi format in the file formats tab Add option to have FastPlay not automatically hide to tray. Add new bookmarks option. Press M to place a bookmark at the current position in the file. Press control M to open the list of bookmarks. Press enter on a bookmark to jump to it, or press delete to remove it. Attempt to fix internet streaming crashes. Add the ability to record the player's direct output. Useful in many applications, recording internet radio streams, recording affected audio, etc. Press R to toggle recording. Set up options in the recording tab. Add a radio dialog, for saving and quickly playing internet radio streams. Search tab coming soon. Thread youtube search, so NVDA doesn't freeze.
#FastPlay 0.4.1: Add toggle record and speak now playing as an available global hotkey assignment Add new speech tab, with some new settings for verbosity. Fix volume over 100 setting Add shuffle Fix a bug with tags where they were getting cutoff in some cases. Adjust volume curve and step. Fix flac playback by actually including the proper version of bassflac. Add home to jump to the beginning of the file. Add J for jump to time.
Are there more plans for additional file formats? I want to put my audible m4b stuff in here so I can read it with tempo. I also don't know if chapters are supported. Lots to drop on you today but I really like this thing lol.
seriously... I had better just keep quiet about the quality, I've said all I can already. The only thing I'm wondering at the moment is whether it's possible to open an entire folder with nexted folders all of which contain music. Is this already doable and, if not, can it be managed at some point?
#FastPlay 0.4.2: Add convolution reverb. Add all files option to the open file dialog. Add chapter seeking option. in the speak now playing key, remove the now playing: text Make it so shift numbers show a text field in a dialog with the tag that that number key would speak. Add open folder option Add playlist manager dialog, control P. Allows you to modify the current playlist. You can add files to it by pasting them, move them with alt up and down arrows, delete them with delete.
Bug report. 1. Start playing a song such as from internet radio. 2. Press number 1. Result: If the song title contains an apostrophe, the screen-reader only speaks up to the apostrophe and nothing more. Example: Mabel - Don, representing, Mabel - Don't Call Me Up.
Just a thought. Do we have a keystroke to mute the player? Ordinarily you might want that but also I'm thinking that if you did implement scheduled recordings, the user may well want the player muted while that was happening.
When you say "fix", does that mean that volume can now go over 100%, or that it can no longer do so? I always like the option of going past 100% for quiet files.
This is a much better release, really nice. Streaming works now especially when adding stations to the Radio dialog. Love the recording feature. I never expected that for a minimal player but a great idea. If you add the ability to schedule recordings to be made at specific times you have one hell of a player there. But if not, it's still wonderful. One suggestion. In Playback tab, where you have the ability to suppress speech output for track changes, can you apply that same setting to the volume level with up and down arrow so it doesn't tell us the volume percentage? I'll be using it a lot throughout the day.
@brian_hartgen I can add scheduling, but the player would have to be running for the schedule to occur. I'm thinking, you add a stream to the radio dialog, and then the scheduler would let you choose dates and times, repeat options, etc, and then which stream to play, duration, etc. Not too sure yet
Interesting bug. Speech Tab. Unchecking volume has no effect. Volume level is still spoken. However, if I uncheck Speak effect value when adjusted, that stops the volume level being spoken.
@brian_hartgen So there are actually two ways to adjust volume. control up and down adjusts volume regardless of what you have set as effect, and that's what the volume setting toggles speech for.
Are we talking about the titles now? Could Shift numbered keys output the required segment to an edit field? Also, if we're not going to get the track in the window title, and I understand why that would be, in speech verbosity, please add option to suppress verbeage such as, Now Playing, when number 1 is used. We know we're playing it.
@brian_hartgen The only thing with that is then NVDA would automatically announce track changes regardless, as it speaks automatically whenever the window title updates, otherwise yes I'd be all for this.
I definitely want to see this improve. Its literally the foobar replacement I could ever want. It just works. I do wish I could turn off volume notifications and maybe shuffle files, but it is very nice so far.
Also something I'm noticing is that it takes lot of volume ticks to actually lower it a lot for me. This could be my new headset, but with foobar I could set volume step and it lowered quickly, even if volume step was set to three only. Fastplay lowers by 5% but it seems to be less? Perhaps percentage is not the same as volume step?
@johann not really. Apple development is pretty hostile. I keep running into issues where FastSM just keeps on crashing, but only on Mac and it's been stupid to fix each time. I don't really wanna deal with that crap again
This is absolutely terrific! Forgive me if I've overlooked something, but would you consider adding a keystroke to move back to the beginning of a file? And would a jump to time option be possible? I completely appreciate this has been designed to be minimalistic.
Initially I thought, meh. But I tried it and this is impressive fast even when playing shit from my HDD, has everything I would want from a minimalist media player. Good. May I ask what it was written in?
Ok good just turned into awesome, I just now understood the concept and this is, strong! Nice. One thing, could we get backspace or something to reset a slider to the initial value. Maybe i missed that feature but would be practical imo.
I downloaded FastPlay this morning, and I'm very pleased with it so far. I'm using it as my primary radio and audio player at the moment. I'm enjoying using fastSM as well. Thanks for developing two such excellent apps.
@jscholes @Scott and then they complain that nobody develops for it, and other people complain that no blind people use it. BTW, I suppose I should congratulate you that it's not a VO issue, if it's not.
If I could, I would go yell it off the roofs, but it's too cold. The Muppet Show is coming back... and it's nice and edgy! Love it! facebook.com/reel/154412354687…
It's almost time to start the show!
The Muppet Show, a special event with guest star Sabrina Carpenter, and special guests Seth Rogen and Maya Rudolph, premieres February 4 on Disney+ and ABC.
@Bri Yeah, I'll have to go review their Muppets talking thingy that I bought a while back. Oz had a disagreement with Disney, so he chose not to participate in anything Muppet related.
Yeah, not a fan of the replacement actors they found. Bert and Ernie aren't Bert and Ernie anymore, Cookie Monster is broken, and Elmo is probably the closest out of all of them but still kinda strange
It funny that on one hand MS does not let you update to Win 11 without tpm for encryption stuff, and other hand stores recovery key on their servers and gives it on demand...
Codeberg and Forgejo will have a stand (K5) at FOSDEM 2026 in Brussels. Come by and chat with Codeberg and Forgejo contributors.
There'll also be Codeberg and Forgejo stickers. If your Codeberg or Forgejo hosted project has stickers and you're looking for a way to distribute them: Get in touch and we can organise it!
New FastSM Release: latest
Add option to collapse consecutive usernames in posts
New setting in Templates tab allows limiting how many @mentions are shown
at the start of a post. Excess usernames are replaced with "and X more".
This is a repository of apps to be used with your F-Droid client. Applications in this repository are official binaries built by the original application developers, taken from their resp. repositories (mostly Github, GitLab, Codeberg).
Hoy he conseguido terminar la prueba de completud de los reales (si una secuencia es Cauchy en R, converge en R) y me ha costado un huevo y parte del otro.
No sé si habría sido más fácil con Dedekind cuts, pero el tema de la diagonalización de la secuencia límite no me acababa de entrar.
Naturally, the third comment is written by a non-screen-reader user and says that the article written by a screen reader user about screen readers is completely wrong.
Would you believe that their "experience is the opposite"? They even have the numbers to prove how wrong the author is! Their solution works at 14,000 words per minute!
clearly we're simply not using enough LLM and OCR. Some nerd on HN thought of this after thinking about the problem extensively for about .5 seconds and determined that screen readers suck and it could all be fixed with those two things.
Winter blue tardis
in reply to Majid Hussain • • •Majid Hussain
in reply to Winter blue tardis • • •Winter blue tardis
in reply to Majid Hussain • • •