@FreakyFwoof Here they worked immediately. But for some people a restart was necessary. And have you placed them, by default of course, in C:\TTS and applied both reg files?
@JamminJerry will be curious how people find lag on these voices. At this point it's probably a more hopeful project for me than this Mindmaker voice, especially as that one has some very oddities in pronouncing things wrong (like "Tim" as "time", "gemini as Gemeeni" and such) - so I'd imagine, yeah, this voice will do better. Ha.
@Tamasg Hmm. It's quite responsive over here on my machine with a great CPU and lots of RAM. But it's definitely a bit laggier than Microsoft non-HD voices. And its pronunciation is great. @JamminJerry @FreakyFwoof
Actually no. The latest Windows 11 dev release only provides these two US English HD voices. But this seems to be the beginning, like the previous high-quality voices. Microsoft may well extend this so-called HD upgrade to other voices in the near future. At this time we just have access to these two offline HD voices, and even these haven't been released to the official channel. The rest are not HD though they are high-quality. @JamminJerry
@JamminJerry I was thinking you could probably combine the 32/64 registry files into one, less work if people need to modify. A single find & replace gets both jobs done, you know?
I was only thinking of putting it there as that is where you are suppose to unzip the other voices, so I just thought keep it simple, and put it there as well.
@JamminJerry So, I installed 32-bit and 64-bit adapters, downloaded Ava and Andrew, stuck them in c:\tts, ran the reg files, but I only seem to have the online voices. Am I missing something obvious?
@BorrisInABox @JamminJerry That was me for a while. I only ran the 64 bit at first, then realised I should fix that. Try restarting NVDA if it still doesn't pick up. It eventually did for me.
@BorrisInABox You can open the installer and uncheck "Enable Microsoft Edge online voices" to get rid of the online voices. The Close button applies any settings, so no need to uninstall and reinstall. You should have at least silent spots where NVDA expects Andrew and Ava, even if the voice data was extracted to the wrong location. Is this not the case for you? You can also press Alt+DownArrow to open the list of voices without actually changing them. @FreakyFwoof @JamminJerry @amir
So, anyone tried this on arm yet? My potentially fastest Windows machine right now is an arm virtual machine running on a Macbook M4. There's an arm adapter, but I don't know how that works. Haven't attempted it yet myself.
Well, this can be extended even further if you are into experiments - LOL! There's a newer, non-official installer of NaturalVoiceSAPIAdapter with Microsoft.CognitiveServices.Speech updated to 1.46.0. Download from Google Drive - link at the end. Delete old versions of NaturalVoiceSAPIAdapter and install the new one. Now Ava and Andrew become more HD-like. And now you can switch between HD and regular voice versions by editing the EnableHDVoice=yes (EnableHDVoice=no) parameter in the 1033.INI file in the voice folder. drive.google.com/drive/folders…
now for my stupid question of the day. I didn't think you could do it, but based on your message, you can, so can you install both? I have the normal NVDA, but I also have the alpha version that is 64 bit.
@JamminJerry Install both, no unless you pick a diffrerent location, but run both reg files, yes. Alpha will pick up on 64, standard will pick up on 32.
I didn't break them! they work! woohoo! it does seem a bit laggy though. I am using ava right now. I saw all of the online voices too, and they seem to work as well, but I was trying to find the offline ones, but I eventually found them. lol. they were above the blasbay studios voice I baught.
this isn't just the lag. there is a slight, how do I explain it, from time to time, when it starts talking there is a very tiny jerk? pause? something like that when it starts talking. it is very small, and it isn't all the time.
@JamminJerry Andrew is less expressive though a lot more than other MS voices. I really like both. Just a shame Microsoft hasn't opened them for use with other screen readers. That's why we should resort to this type of so-called data mining of sorts.
you ask, and I shall deliver. this took me several times to record though. one time I didn't have the right button pushed on the mixer. another time I had more than just the NVDA channel in the recording. one time I didn't have any channels selected, so it was just a silent file, but after several attempts, here you go. I personally think there should be a bit more of a pause at the between paragraphs, but that is just me.
sadly I can't edit the post in tweesecake, because I have found if I do, it removes the attachment. I would have to remove it, and just do a new reply.
@JamminJerry Funny thing, Ava on my machine is lower pitched even though it's set to 50, and sounds quite a bit different than in your recording. Strange times.
people are probibly going to disagree with me, but at least to me, this voice sounds a little bit like the cortona voice. like I said, this is just me though. maybe I am way off.
@JamminJerry I'm beginning to wonder whether it's actually working properly. Both Andrew and Ava are in my case, Ivona voices, so though the name says one thing, the voice is another, which was already installed on my system.
@JamminJerry I think it's not using that voice at all, like I say. It's using some other voice on the system already, and claiming it's Andrew or Ava. I can't prove it apart from on my own system, but compare the sound of the voices you have, to things already on your system.
In the first draft of the Black Mirror episode for this scenario, a Waymo exec has a medical crisis and the ambulance can't get there on time because the streets are choked with Waymos.
In the final script, Waymo invites PG&E employees to their blowout holiday party, with free Waymo pickups. Most of the staff is in those Waymos when the power goes out; they're stuck all over the city, unable to solve the problem.
SF’s is having a massive power outage. We ended up having to drive across SF to find a laundromat to dry our soaking wet laundry. Here’s my two observations: every third block is a traffic jam caused by a confused Waymo that doesn’t know how to handle these intersections with no power. Why are the Waymo’s even allowed to drive right now?
Also, considering how much of our city’s budget goes to the cops, why aren’t any of them directing traffic? I mean, it’s not like SFPD is busy solving crime…
SF’s is having a massive power outage. We ended up having to drive across SF to find a laundromat to dry our soaking wet laundry. Here’s my two observations: every third block is a traffic jam caused by a confused Waymo that doesn’t know how to handle these intersections with no power. Why are the Waymo’s even allowed to drive right now?
Also, considering how much of our city’s budget goes to the cops, why aren’t any of them directing traffic? I mean, it’s not like SFPD is busy solving crime…
Power outage in most of San Francisco, including traffic lights. Autonomous cars going nuts. I shall stay home (in the dark) in case I get run over by a Waymo.
Long discussion on the orange website about the NIST's Boulder NTP site being down. Not a lot of people with practical experience in the topic.
NTP is designed to function as a tree model, rooted in "stratum 0" NTP servers with direct time references. Actual clients are expected to interact with lower-stratum servers that themselves synchronize to higher-stratum servers. Generally any NTP server with a GPS time reference is considered suitable for stratum 0, and the vast, vast majority of real-world NTP/PTP deployments are ultimately working off of one or more GPS references. GPS operates from the Naval Observatory clock via a few different layers of indirection, each of which has its own high-precision oscillators to allow plenty of holdover. Purpose-built stratum 0 servers, used e.g. in data centers, have their own precision reference oscillators for holdover should the GPS lock be lost. Basically any real problem in this space would probably evolve pretty slowly.
The NIST NTP servers have always suffered from high load, to the extent that they've sometimes been unresponsive in the past. Synchronizing devices directly to these NTP servers is a bad practice, discouraged by the NIST and now mostly gone from default configs. That limits the impact of an outage a lot. The authenticated service is intended to help with the load-related reliability issues but presumably its users are sophisticated enough that their stratum-1 will be fine. Most clients are interacting with lower-stratum NTP servers operated by, for example, ntppool.org. These track multiple stratum-0 references and will be fine with the loss of one of the NIST sites.
There are probably still a bunch of stupid clients that try NIST Boulder NTP servers exclusively (I know some old network appliances shipped like this by default), and they might be a little driftier than usual for a while, but I wouldn't expect any of these to be important as they were pretty much already misconfigured.
Big picture, the NTP system is very big and very distributed and the NIST Boulder site has no real significance beyond name recognition---the same name recognition that made it a poor choice for most purposes.
Hey open source folks that have large repositories and contributors on their projects, is GitHub still a safe and viable option? What alternatives are worth considering? And are the pipelines available well tuned and compatible to build flatpacks and snaps? #askfedi #openspurce #github
I use GitLab for my projects, their CI works well enough but only if you host your own runners. GitLab's atrociously garbage free CI minutes mean that CI as a merge request validation step is basically impossible, as merge requests' CI runs on the submitter's account and runners not your project's runners and if the submitter doesn't have any CI minutes (they almost certainly won't) then the jobs don't run. It used to be awesome before they nerfed their CI into the ground.
yeah I was super vague, sorry. Decently sized in terms of community engagement and participation for issues and PRs, with good reviewing and moderation tools. "Safe" as in avoiding things like "we're now charging you 10 times more for runners" or "we're sponsoring ICE". I know the second point is mostly impossible to account for in perpetuity
This is how I feel about Firefox too. There's no alternative. No, really, Librewolf is Firefox, who exactly do you think you're sticking it to? And Vivaldi is a proprietary browser built on Google's engine, you're sticking it to nobody but yourself. Oh, you're switching to Servo? Not this decade you're not.
Mozilla CEOs have always been incompetent, but there's no other alternative outside of the direct control of people who are many orders of magnitude worse. We support Firefox or we lose the web. mastodon.gamedev.place/users/D…
I don't think I'll stop using Firefox anytime soon. Yes, their management are fucking idiots and it's annoying that I'll have to disable new AI features whenever they release them..
But the alternative is to use something Chromium-based, which would make Google's domination of web technology absolute - and Google is 100x more evil than Mozilla ever could be.
I hope this fucking bubble pops before Mozilla fucks up Firefox so badly that it becomes completely unusable
Hey #Mastodon people, the #blogging hashtag is pretty cool because I get to see blog posts from folks I would otherwise not meet. Can you recommend any other hashtags where people share their essays or fiction?
PS: I don't really want a hashtag where people share their thoughts about writing and such. I am inundated with those lol. I want actual pieces so I can find new folks to read ✌️
People from out of town who are vaguely into tech are always excited about coming to see the campuses of the big tech companies they know. It’s hard to describe to them how it’s pretty much like coming to see a parking lot, it’s not that exciting, it’s very devoid of life and energy by design
"Money" is a perfectly valid response to "why do you want this job?". Sick of interviewers expecting me to act like I've dreamt of working at Morrisons since I was a child.
Very, very Early Mindmaker FlexVoice NVDA-addon. It does not support speech rate. it does not support many things. it does support voice. no indexing yet. I'm still enhancing the wrapper. But you know what? We do get speech. So it's a start. eurpod.com/flexvoice.nvda-addo… if you want to test it and a broken synthesizer as it evolves
@BorrisInABox @fireborn ahaha yeah it does that sometimes when there's too small chunks of audio. best works if you start it from a notepad window, then for some reason it can lock after first chunk better. Still investigating that one.
The thing I most remember about Flexvoice when it was still a thing was thinking "wow, I would never want to actually use this as a speech synthesizer", so it will be interesting to see what comes of this in an environment where I actually can.
We backed up Spotify (metadata and music files). It’s distributed in bulk torrents (~300TB). It’s the world’s first “preservation archive” for music which is fully open (meaning it can easily be mirrored by anyone with enough disk space), with 86 mil…
Happy Holidays from Pneuma Solutions! Have some completely free #remoteIncidentManager usage on us! From now through January 2nd, we're inviting you to try the full run of Remote Incident Manager's capabilities free of charge! Download RIM today and …
Windows tablet users that use NVDA, is there a a way that isn't convoluted to get it to behave in a way that makes sense? e.g. Swiping up, down, left, right for arrows, left and right swipes with 2 fingers for shift tab and tab, double tap to activate whatever keyboard focus is on, etc. Object nav seems silly when this would work considerably better.
So, a random pet peeve of mine: Listening to the football game and the anouncer just said, "25 seconds left to go". Take your pick. "25 seconds left", or "25 seconds to go". "25 seconds left to go"? just, no!!!
"Today, as other browsers race to build AI that controls how you experience the web, we are making a clear promise:
We’re taking a stand, choosing humans over hype, and we will not turn the joy of exploring into inactive spectatorship. Without exploration, the web becomes far less interesting."
Browsing should push you to explore, chase ideas, and make your own decisions. It should light up your brain. Vivaldi is taking a stand. We choose humans over…
Essa tal coisa que é a vida social também está a ficar toxica, cfedo. Tive o asar de comentar com uma pessoa que se diz de esquerda o facto de ter simpatia pessoal pelo amarques Mendes. O drama! O Horror! Como é que podes! O conceito do “pessoal” parece que deixou de funcionar.
@catarinac De quem? Só se for dos próprios… Eu conheço o tipo das festas da escola, era simplesmente mais simpático e afavel que os restantes pais betinhos dos meus colegas. Ninguem vota num gajo por isso.
Come to terms with this: you are owned. You are a slave. I am a slave. We're all slaves. Our owners are the ones whose names appear in newspapers, and on our ballots, and on our receipts.
Understand that you are bought and sold, and your wishes are as relevant to public policy as my cats' opinions on free software.
Only when you begin to understand this (and that I am not joking in any way at all, I am 100% deadly serious about this) will you grasp how fucked we all are, and will be, until we either die or fix it.
The reason you and I are allowed to consume resources and to earn and spend money is to produce goods and services for the landed elite. Your raison d'etre is to create value for capital. They give you just enough to survive on without having us rise up and kill them all. There is only land for you to live on because they permit it.
If the rich could get everything built and served by robots and AI, they would happily kill everyone else alive to take everything for themselves. This is pretty clearly the end game if you think about it for a minute.
If you wonder about AI regulation, look at what was done for tobaco, asbestos, PFOA (aka forever chemicals) and many other dangerous products. And also about greenhouse gas emissions.
We are happy to announce that GUADEC 2026 will take place from July 16th to 21st, 2026, in A Coruña, Spain. As in recent years, GUADEC will be held as a hybrid event, allowing attendees to participate either virtually or in person at our venue.
Andre Louis
in reply to Amir • • •Amir
in reply to Andre Louis • • •Amir
in reply to Andre Louis • • •Andre Louis
in reply to Amir • • •Amir
in reply to Andre Louis • • •Andre Louis
in reply to Amir • • •Amir
in reply to Andre Louis • • •Andre Louis
in reply to Amir • • •JamminJerry
in reply to Andre Louis • • •Tamas G
in reply to JamminJerry • • •Amir
in reply to Tamas G • • •JamminJerry
in reply to Andre Louis • • •Amir
in reply to JamminJerry • • •drive.google.com/drive/folders…
@FreakyFwoof
AndrewHD and AvaHD – Disk Google
Google DriveAndre Louis
in reply to Amir • • •Amir
in reply to Andre Louis • • •@JamminJerry
Andre Louis
in reply to Amir • • •JamminJerry
in reply to Amir • • •Andre Louis
in reply to JamminJerry • • •JamminJerry
in reply to Andre Louis • • •Andre Louis
in reply to JamminJerry • • •Borris
in reply to Andre Louis • • •Andre Louis
in reply to Borris • • •Borris
in reply to Andre Louis • • •Borris
in reply to Borris • • •Timothy Wynn
in reply to Borris • • •Andre Louis
in reply to Borris • • •Borris
in reply to Andre Louis • • •Borris
in reply to Borris • • •Andre Louis
in reply to Borris • • •Amir
in reply to Borris • • •@FreakyFwoof @JamminJerry
Borris
in reply to Borris • • •Amir
in reply to Andre Louis • • •There's a newer, non-official installer of NaturalVoiceSAPIAdapter with Microsoft.CognitiveServices.Speech updated to 1.46.0. Download from Google Drive - link at the end.
Delete old versions of NaturalVoiceSAPIAdapter and install the new one. Now Ava and Andrew become more HD-like.
And now you can switch between HD and regular voice versions by editing the EnableHDVoice=yes (EnableHDVoice=no) parameter in the 1033.INI file in the voice folder.
drive.google.com/drive/folders…
AndrewHD and AvaHD – Disk Google
Google DriveJamminJerry
in reply to Andre Louis • • •Andre Louis
in reply to JamminJerry • • •JamminJerry
in reply to Andre Louis • • •Andre Louis
in reply to JamminJerry • • •Amir
in reply to JamminJerry • • •JamminJerry
in reply to Amir • • •Amir
in reply to JamminJerry • • •JamminJerry
in reply to Amir • • •Amir
in reply to JamminJerry • • •JamminJerry
in reply to Amir • • •Amir
in reply to JamminJerry • • •JamminJerry
in reply to Amir • • •Amir
in reply to JamminJerry • • •JamminJerry
in reply to Amir • • •Amir
in reply to JamminJerry • • •JamminJerry
in reply to Amir • • •Amir
in reply to JamminJerry • • •JamminJerry
in reply to Amir • • •Amir
in reply to JamminJerry • • •JamminJerry
in reply to Amir • • •Andre Louis
in reply to JamminJerry • • •JamminJerry
in reply to Andre Louis • • •Andre Louis
in reply to JamminJerry • • •Andre Louis
in reply to Andre Louis • • •Amir
in reply to Andre Louis • • •JamminJerry
in reply to Amir • • •Andre Louis
in reply to JamminJerry • • •Andre Louis
in reply to Andre Louis • • •Amir
in reply to Andre Louis • • •@JamminJerry
Andre Louis
in reply to Amir • • •Amir
in reply to Andre Louis • • •Andre Louis
in reply to Amir • • •Amir
in reply to Andre Louis • • •JamminJerry
in reply to Amir • • •Andre Louis
in reply to JamminJerry • • •