The latest versions of Common Voice datasets, Scripted Speech v27.0 and Spontaneous Speech v5.0, are available for download on the Mozilla Data Collective! A giant thank you to all of our contributors for their hard work putting together these datasets.

What’s new in this release:

Version 27.0 of Scripted Speech includes datasets for 295 languages. These datasets contain 32,349,999 voice clips, making approximately 42,593 hours of speech data available for developing and improving speech technology.
Version 5.0 of Spontaneous Speech includes datasets in 80 languages. These datasets contain 89,754 voice clips of free-form answers to questions, of which 302 hours have been transcribed and validated.

Since the last release, the Common Voice community has welcomed 1 language to Scripted Speech - Pa’O (blk) - and 6 languages to Spontaneous Speech - Chinese (China, zh-CN), Swahili (sw), Palauan (pau), Sundanese (su), Bengali (bn) and Shan (shn).

“DoNotSleep”

wakeLock.acquire() xD

en.wikipedia.org/wiki/Pictures_for_Sad_Children

In some ways “outdated”, too edgy, but in some details maybe even more relevant today than it originally was.

(1/3 of humanity having no electricity is a bit exaggerated, in 2000 it was 1/5.)

JoJo no Kimyou na Bouken: Steel Ball Run - 2nd - 3rd STAGE, episode 1

Reminder: Please do not discuss plot points not yet seen or skipped in the show. Failing to follow the rules may result in a ban.

All discussions

This post was created by a bot. Message the mod team for feedback and comments. The original source code can be found on GitHub.

Pick it, pack it, fire it up, come along
and take a hit from the bong.

Ganso! Bandori-chan, episode 51

Reminder: Please do not discuss plot points not yet seen or skipped in the show. Failing to follow the rules may result in a ban.

All discussions

This post was created by a bot. Message the mod team for feedback and comments. The original source code can be found on GitHub.

Ninja Theory - developer of the Hellblade series - is shutting down as part of today’s Xbox cuts. Its proposed buyers pulled out, resulting in the closure.

Good news, a security issue that was blocking the draft for the new DNS-PERSIST-01 challenge mode for TLS certificates has been resolved, meaning the challenge can resume moving forward.

Some background for those who aren’t up to date:

• TLS challenges are how self-hosters and many other TLS certificate users prove who they are in order to automate cert renewals.
• The most popular challenge methods are HTTP-01 (which involves having a live web server answering on port 80 somewhere) and DNS-01 (which involves putting a key given by the TLS authority in your DNS zone in a TXT record temporarily).
• All the current challenges are conducted “live” and require active online participation on the part of the domain owner each time a cert is approaching expiration.
• The new DNS-PERSIST-01 challenge would let owners keep a proof in their DNS zone indefinitely (or with self-imposed expirations), eliminating the need for those online sessions.
• DNS-PERSIST-01 was slated to come out in Q2 this year at the latest and all the popular tools (ACME bots, reverse proxies etc.) have been chomping at the bit to implement it.
• Unfortunately there’s been a security snag. Trying to do things the way DNS-01 did (record only the server-provided proof in DNS) would open DNS-PERSIST-01 to middleman attacks, because unlike DNS-01 where the proof is only there temporarily for a few seconds or a few minutes at the most, in this case the proof would be long-term.
• Hence issue #64, which has been trying to come up with a way to keep only a partial proof in DNS, and combine it with something issued by the server and something issued by the domain owner to make the whole proof, thus rendering the stuff stored in DNS impossible to use for malicious purposes, yet still sufficient for verifying ownership periodically.

Hopefully now that this hurdle has been passed things will proceed faster and we’ll get to use this in the near future.

Manufacturing custom chips has become easier and cheaper during recent years. wafer.space is not the only one doing so.

They use a 180nm, which is meh. But still, it’s an advancement towards making this accessible to everyone. Maybe in the coming years, we could see people easily doing so at home.

Also now, IMO, importance of open source hardware design has increased. 20 years ago, it wouldn’t mean much if a CPU is open source or not. You could not produce a chip. And a discrete one is useless. Now it’s different.

NoteTrace is a self-hosted alternative to Google Keep, Evernote and Apple Notes: notes, checklists and reminders in a card grid, on your own server. AGPL-3.0, a single Docker container, a web app that works offline, a native Android app, and a Wear OS app. No telemetry, no cloud. This is the first stable release.

Part of the TraceApps family: NutriTrace (nutrition), CookTrace (recipes / pantry / shopping), LiftTrace (strength / lifting).

What it does

• Notes and checklists. Pins, colors, nested labels, archive, trash, version history, and full-text search that also finds words in voice transcripts and in photos.
• Reminders and Tasks. Repeating reminders that keep their local time, exact alarms on Android, and one view of everything that’s due.
• Offline everywhere. The installed web app keeps editing with no connection and syncs later; Android and the watch work offline too.
• Wear OS. Tick off lists, read notes, see what’s due, and speak a note (“call the plumber tomorrow at nine” sets the reminder), with a tile and a watch face complication.
• Voice notes and drawings. Record with the screen off and get a transcript you can tap to jump to.
• Bring your notes with you. Import from Google Keep, Evernote, Memos, Blinko and Markdown vaults; export everything as Markdown with images.
• Optional AI. Tidy up, summarise, or turn a note into a checklist, with the provider of your choice. Off unless you set it up.

Links

• Repo: github.com/TraceApps/notetrace
• Docs: traceapps.github.io/docs/notetrace/
• Release notes + signed APKs (phone and watch): github.com/TraceApps/notetrace/releases/…/v1.0.0
• Docker: ghcr.io/traceapps/notetrace:latest (mirror: traceapps/notetrace:latest)

AI Disclosure

Per Rule 7 / [AIP] disclosure requirements AI was used during development as a coding assistant. Level per category:

• Design (architecture, system design): Hint: I make the architectural calls; AI suggests trade-offs and edge cases I might have missed.
• Implementation (production code): Pair: roughly 50/50. AI drafts, I review, adjust, test on real hardware, and only commit what I’ve verified. Every commit is manually reviewed before it goes to my dev repo.
• Testing (writing tests, test plans, QA): Assisted: real-device testing is manual (I test on my own PC and mobile devices before every release). AI helps draft test plans and think through edge cases.
• Documentation (docs, comments, README, CHANGELOG): Pair: release notes and changelog entries are drafted with AI then edited for tone; comments and code docs are mostly Pair as well.
• Review (code review, PR feedback): Assisted: I’m the reviewer; AI helps with security sweeps, audit passes on complex changes, and consistency checks.
• Deployment (CI/CD config): Hint: Docker/GitHub Actions/release pipeline is largely conventional; AI-suggested improvements only.

cross-posted from: lemmy.world/post/52212601

Source

idk, bad pun I just came up with. btw saifu (財布) = wallet in Japanese

cross-posted from: https://piefed.world/c/tech/p/1421710/early-rogue-ai-agent-activity-and-attempts-to-hack-found-on-urlquery-net

Artist: Orouu | pixiv | twitter | linktree | danbooru

Full quality: .webp 1 MB (991 × 1674)

Youjo Senki II, episode 12

Reminder: Please do not discuss plot points not yet seen or skipped in the show. Failing to follow the rules may result in a ban.

All discussions

This post was created by a bot. Message the mod team for feedback and comments. The original source code can be found on GitHub.

Tsuihou Sareta Tensei Juukishi wa Game Chishiki de Musou Suru, episode 13

Reminder: Please do not discuss plot points not yet seen or skipped in the show. Failing to follow the rules may result in a ban.

All discussions

This post was created by a bot. Message the mod team for feedback and comments. The original source code can be found on GitHub.

also my teto plush committed suicide

cross-posted from: https://piefed.world/c/tech/p/1417255/young-users-ditch-google-for-ai-with-unknown-consequences

According to the officials who spoke to Bloomberg, some personnel inside U.S. Central Command leaned too hard on [Palantir] to accelerate intelligence analysis and decision-making. The Defense Department has reportedly made the software tool a cornerstone of military operations over the past year… Officials said some users expected Maven to flag stale records or contradictions in the intelligence assembled for potential targets, though it is not clear why they thought the system would do that. The Minab [elementary school massacre] site, which was cataloged as an Islamic Revolutionary Guard Corps facility due to outdated data, was fed into Maven with other candidates and came out as a recommended day-one target. Target-list work that once took hours was condensed into minutes.

A Palantir spokesperson told Bloomberg the company “is not responsible for the underlying data nor identifying intelligence deficiencies” and that there is no evidence its software was at fault. Two people familiar with Palantir’s Pentagon contracts said the government keeps primary responsibility for the quality of the information that’s fed into Maven.

cross-posted from: lemmy.world/post/52254714

It will be okay

cross-posted from: lemmy.blahaj.zone/post/48313374

Promotional

Im not affiliated to Fermi/Harmony chat in any way, im just a user who liked it and want to share my experience of the app i find worthy of sharing

so i wanted to share a promising project that I have been using for a few months.

made this post to promote and share this platform since it needs more people, theres arent many atm and it deserves way more.

Fermi/Harmony Chat is free and open source discord like guild/community centered chat platform with most of the features that are expected from those platforms like message channels, vocal chats, threads, screensharing and more. Importantly the chat platform has a strict stance against LLMs AIs and AI generated slop like vibe coding and other abusive usage of AI that replaces human work and slop it in the process. sadly fermi/harmony is unique of having a strong stance against AI. (other similar projects sadly uses AI and excuses AI usage and vibe coding)

the platform also allows users to create their own client for the platform. harmony chat is the API and fermi is just the currently only working client on the platform (was a spacebar client previously until harmony was created as a spacebar fork with the goal of fixing the many issues and bugs present on spacebar) and in the future, instances are going to be federated and the process will be managed directly by the client (fermi and other clients) without being tidious and for advanced users like on matrix, so more easy and beginner friendly.

the platform also is more lenient on its acceptable policy and have a no bullshit moderation. (sadly not the case for other alternatives which sucks very much)

IMPORTANT

currently the project in its early stage and still has lots of bugs, glitches and various issues.

it is currently only accessible through a web browser (there is no app neither on desktop or on mobile)

lots of features are not finished and the platform still need major improvements. but i hope it is going to grow and become a viable alternative.

currently there are around 800 people are registered on fermi/harmony chat.

fermi.chat the main website currently (use this one to register) melodychat.org the organisation’s website (its called melody on codeberg and on the website since harmony was taken)

codeberg link of both harmony and fermi :

codeberg.org/MelodyChat/Harmony codeberg.org/MelodyChat/Fermi

credit to Mathium5, eliza, devs and contributors for developping and maintaining fermi and harmony chat.

cross-posted from: ani.social/post/37285826

Alt:

Sourced from Pixiv. Absolutely adore how good she looks but upon closer inspection you can see nips so I didn’t think it would fly over on sh.itjust.works/c/rwby (but please come join for more RWBY fanart & discussion <3 )

I am in the hunt right now for a domain (I plan to use for email and such, meaning, things that would be tied to my legal person). As some of you know, by law in many places it is required to have a set of legally true information to register a domain (from name to house address, in case the domain needs to be verified to belong to a real person).

I was checking porkbun a second ago, and the privacy policy really does not give me confidence (they mention very clearly there is sale and ai training involved). I am from Europe so I would expect gdpr to be respected by default, however, they say I would have to send an email to get my rights and identify myself (I expect an Id upload which is not really a thing I want to do).

Considering all this, what are the better places for domains? I expect to pay more than porkbun, but if it people that do it right, so be it.