<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Jeson</title>
    <description>The latest articles on DEV Community by Jeson (@jeson_image2img).</description>
    <link>https://dev.to/jeson_image2img</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4216471%2F0ff73a31-aa67-4f12-8de4-9d670e98328d.png</url>
      <title>DEV Community: Jeson</title>
      <link>https://dev.to/jeson_image2img</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/jeson_image2img"/>
    <language>en</language>
    <item>
      <title>I made a free browser version of the "word impostor" party game — pass-and-play or online rooms</title>
      <dc:creator>Jeson</dc:creator>
      <pubDate>Tue, 10 Nov 2026 00:09:59 +0000</pubDate>
      <link>https://dev.to/jeson_image2img/i-made-a-free-browser-version-of-the-word-impostor-party-game-pass-and-play-or-online-rooms-4ik2</link>
      <guid>https://dev.to/jeson_image2img/i-made-a-free-browser-version-of-the-word-impostor-party-game-pass-and-play-or-online-rooms-4ik2</guid>
      <description>&lt;p&gt;At game nights my friends kept playing "Who's the Impostor?" with slips of paper. Someone always peeked, and someone always lost the slips. So I built &lt;a href="https://www.imposterwords.com/" rel="noopener noreferrer"&gt;Word Impostor&lt;/a&gt;, a free browser version you can play on one phone or across everyone's phones.&lt;/p&gt;

&lt;h2&gt;
  
  
  How it plays
&lt;/h2&gt;

&lt;p&gt;3 to 10 players. Most players get the same secret word; one player — the Impostor — gets a related but different word. Everyone takes turns describing their word without saying it, then you vote on who seems off. The Impostor wins by blending in.&lt;/p&gt;

&lt;p&gt;For bigger groups (6+) there's an optional &lt;strong&gt;Mr. White&lt;/strong&gt; role who gets no word at all and has to bluff from context.&lt;/p&gt;

&lt;h2&gt;
  
  
  Two ways to play
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Pass-and-play&lt;/strong&gt;: one device goes around the table; each player taps to see their word privately.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Online rooms&lt;/strong&gt;: the host creates a room and shares a short code; friends join from their own phones. No app and no sign-up.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Word packs cover themes like food, travel and internet culture.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why browser-only
&lt;/h2&gt;

&lt;p&gt;Party games die the moment someone has to install an app or make an account. Keeping it as a plain web page means anyone at the table can join in seconds.&lt;/p&gt;

&lt;h2&gt;
  
  
  Try it
&lt;/h2&gt;

&lt;p&gt;Grab some friends and open &lt;a href="https://www.imposterwords.com/" rel="noopener noreferrer"&gt;imposterwords.com&lt;/a&gt;. Suggestions for new word packs are very welcome in the comments.&lt;/p&gt;

</description>
      <category>showdev</category>
      <category>gamedev</category>
      <category>webdev</category>
      <category>javascript</category>
    </item>
    <item>
      <title>I built a cat-themed logic puzzle for the browser — no ads, no account, a new board every day</title>
      <dc:creator>Jeson</dc:creator>
      <pubDate>Tue, 10 Nov 2026 00:09:39 +0000</pubDate>
      <link>https://dev.to/jeson_image2img/i-built-a-cat-themed-logic-puzzle-for-the-browser-no-ads-no-account-a-new-board-every-day-38dp</link>
      <guid>https://dev.to/jeson_image2img/i-built-a-cat-themed-logic-puzzle-for-the-browser-no-ads-no-account-a-new-board-every-day-38dp</guid>
      <description>&lt;p&gt;I like Star Battle and Queens-style puzzles, so I made my own take with cats: &lt;a href="https://meowdoku.page/" rel="noopener noreferrer"&gt;Meowdoku&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  The rules (30 seconds)
&lt;/h2&gt;

&lt;p&gt;Place exactly one cat in every row, every column and every colored region. Cats can't touch each other — not even diagonally. That's it. The fun is in the deductions: a small region forces a cat, which blocks its neighbors, which forces the next region.&lt;/p&gt;

&lt;h2&gt;
  
  
  What's in it
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;300 levels&lt;/strong&gt;: Easy 7×7, Hard 8×8 and Expert 9×9.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;A daily puzzle&lt;/strong&gt; that resets at 00:00 UTC, with a global leaderboard.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;A built-in solver&lt;/strong&gt; for when you're truly stuck (or want to check a board you found elsewhere).&lt;/li&gt;
&lt;li&gt;Works on phones with touch controls; progress saves on your device.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Design choices
&lt;/h2&gt;

&lt;p&gt;I wanted it to feel like a paper puzzle book, not a mobile game, so there are &lt;strong&gt;no ads, no account and no download&lt;/strong&gt;. It's a static web app, so it loads fast and works fine on slow connections. Every level is checked to have a unique solution, so you never have to guess.&lt;/p&gt;

&lt;h2&gt;
  
  
  Play
&lt;/h2&gt;

&lt;p&gt;Today's board is waiting at &lt;a href="https://meowdoku.page/" rel="noopener noreferrer"&gt;meowdoku.page&lt;/a&gt;. If you try the Expert levels, tell me which ones felt unfair — I'm tuning difficulty and would love the feedback.&lt;/p&gt;

</description>
      <category>showdev</category>
      <category>gamedev</category>
      <category>javascript</category>
      <category>puzzle</category>
    </item>
    <item>
      <title>What is a "playable world model"? A plain-language guide to Genie 4 and the alternatives you can try today</title>
      <dc:creator>Jeson</dc:creator>
      <pubDate>Tue, 10 Nov 2026 00:09:19 +0000</pubDate>
      <link>https://dev.to/jeson_image2img/what-is-a-playable-world-model-a-plain-language-guide-to-genie-4-and-the-alternatives-you-can-3dhd</link>
      <guid>https://dev.to/jeson_image2img/what-is-a-playable-world-model-a-plain-language-guide-to-genie-4-and-the-alternatives-you-can-3dhd</guid>
      <description>&lt;p&gt;Video generators give you a clip to watch. &lt;strong&gt;World models&lt;/strong&gt; go one step further: they generate a scene you can step into and control, frame by frame, in response to your inputs. That's why people call them "playable worlds."&lt;/p&gt;

&lt;p&gt;I kept losing track of which of these systems were actually usable, so I built &lt;a href="https://playableworldai.com" rel="noopener noreferrer"&gt;PlayableWorld AI&lt;/a&gt;, a small independent hub that tracks them with links to official sources. Here's the short version of what I've learned.&lt;/p&gt;

&lt;h2&gt;
  
  
  Google DeepMind Genie 4 and Project Genie
&lt;/h2&gt;

&lt;p&gt;Genie 4 is DeepMind's latest world model. Right now it's a research preview with limited, invite-only access. Project Genie is the Google Labs experience built on top of it, tied to a paid subscription tier. There's no public API yet, so if you're a developer the honest answer is "watch and wait." I keep the current status and eligibility notes on the site and update them when Google changes anything.&lt;/p&gt;

&lt;h2&gt;
  
  
  Alternatives you can use now
&lt;/h2&gt;

&lt;p&gt;While Genie gets the headlines, several teams ship world-generation tools you can try today:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;World Labs Marble&lt;/strong&gt; generates explorable 3D worlds from images or text.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Tencent HunyuanWorld&lt;/strong&gt; is an open model for generating 3D scenes.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Skywork Matrix-Game&lt;/strong&gt; focuses on interactive, game-like world generation.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Odyssey&lt;/strong&gt; works on interactive video you can steer in real time.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Each has different trade-offs in latency, control and output format (interactive video vs. exportable 3D). Prices change often, so I check them against the providers' own pages rather than repeating numbers from social media.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why this matters for devs
&lt;/h2&gt;

&lt;p&gt;If world models get fast and cheap enough, they could change prototyping for games, robotics simulation and training data. Today they're best for experiments and demos — but it's a good time to learn the vocabulary.&lt;/p&gt;

&lt;h2&gt;
  
  
  Follow along
&lt;/h2&gt;

&lt;p&gt;Everything is collected at &lt;a href="https://playableworldai.com" rel="noopener noreferrer"&gt;playableworldai.com&lt;/a&gt;, including official DeepMind demos and an optional one-email notice when a Genie API opens. Corrections and tool suggestions are very welcome in the comments.&lt;/p&gt;

</description>
      <category>ai</category>
      <category>machinelearning</category>
      <category>gamedev</category>
      <category>news</category>
    </item>
    <item>
      <title>How I transcribe videos without uploading the video (browser-side audio extraction)</title>
      <dc:creator>Jeson</dc:creator>
      <pubDate>Tue, 10 Nov 2026 00:08:56 +0000</pubDate>
      <link>https://dev.to/jeson_image2img/how-i-transcribe-videos-without-uploading-the-video-browser-side-audio-extraction-221e</link>
      <guid>https://dev.to/jeson_image2img/how-i-transcribe-videos-without-uploading-the-video-browser-side-audio-extraction-221e</guid>
      <description>&lt;p&gt;Most online transcription tools ask you to upload the whole video. For a 1 GB screen recording that means a long wait, a lot of bandwidth, and handing your raw footage to a server you don't control. When I built &lt;a href="https://voxcribe.net" rel="noopener noreferrer"&gt;Voxcribe&lt;/a&gt;, I wanted to fix exactly that.&lt;/p&gt;

&lt;h2&gt;
  
  
  The idea: only send the audio
&lt;/h2&gt;

&lt;p&gt;A video file is mostly picture data. Speech recognition only needs the audio track. So Voxcribe pulls the audio out &lt;strong&gt;in your browser&lt;/strong&gt; and sends only that (much smaller) audio for transcription. The video itself never leaves your device.&lt;/p&gt;

&lt;p&gt;In practice this means:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;A 30-minute MP4 that weighs hundreds of MB turns into a few MB of audio.&lt;/li&gt;
&lt;li&gt;Uploads finish in seconds instead of minutes on slow connections.&lt;/li&gt;
&lt;li&gt;Sensitive footage (client calls, internal demos) stays local.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  What you get back
&lt;/h2&gt;

&lt;p&gt;Every line in the transcript has a timestamp. Click a line and the player jumps to that moment, so you can fix names, product terms or accents while you listen. Language is detected automatically, and if the guess is wrong you get a free re-run.&lt;/p&gt;

&lt;p&gt;Supported inputs: MP4, MOV, WEBM, MKV, AVI for video and MP3, M4A, WAV, AAC, OGG, FLAC for audio.&lt;/p&gt;

&lt;p&gt;Export is TXT for free. Paid plans add subtitle-ready SRT/VTT and DOCX.&lt;/p&gt;

&lt;h2&gt;
  
  
  Free tier, honestly
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Your first file (up to 10 minutes) works with no account.&lt;/li&gt;
&lt;li&gt;A free account gets 120 minutes a month.&lt;/li&gt;
&lt;li&gt;Uploaded audio is deleted after 7 days.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Who it's for
&lt;/h2&gt;

&lt;p&gt;I use it for meeting notes, turning tutorials into blog drafts, and making subtitles for short clips. If you record interviews, podcasts or lectures, the edit-while-you-listen view is where most of the time savings come from.&lt;/p&gt;

&lt;h2&gt;
  
  
  Try it
&lt;/h2&gt;

&lt;p&gt;Drop a file at &lt;a href="https://voxcribe.net" rel="noopener noreferrer"&gt;voxcribe.net&lt;/a&gt; — no install. I'd love feedback on accuracy for your language and on the editor; comments here are welcome.&lt;/p&gt;

</description>
      <category>showdev</category>
      <category>webdev</category>
      <category>javascript</category>
      <category>productivity</category>
    </item>
    <item>
      <title>How to Turn Any Photo into a New AI Image for Free (Image-to-Image Guide)</title>
      <dc:creator>Jeson</dc:creator>
      <pubDate>Fri, 30 Oct 2026 05:26:13 +0000</pubDate>
      <link>https://dev.to/jeson_image2img/how-to-turn-any-photo-into-a-new-ai-image-for-free-image-to-image-guide-3in2</link>
      <guid>https://dev.to/jeson_image2img/how-to-turn-any-photo-into-a-new-ai-image-for-free-image-to-image-guide-3in2</guid>
      <description>&lt;p&gt;Most AI image tools start from a blank box and a text prompt. That works for scenes invented from scratch. It falls apart when you already have a photo — a selfie, a landscape, an old family print — and only want to change the style, the hair, or the damage on the print.&lt;/p&gt;

&lt;p&gt;That second job is &lt;strong&gt;image-to-image&lt;/strong&gt;. The model starts from your picture, keeps the person, the pose, and the composition, and changes only what you ask for. Below is a plain guide to how that works, plus a step-by-step run on &lt;a href="https://image2img.net" rel="noopener noreferrer"&gt;image2img.net&lt;/a&gt;, a free browser tool built for exactly this.&lt;/p&gt;

&lt;h2&gt;
  
  
  What image-to-image actually does
&lt;/h2&gt;

&lt;p&gt;Text-to-image tools invent a scene from words. Image-to-image tools take an existing photo as the base. You keep the identity and layout; you steer the look.&lt;/p&gt;

&lt;p&gt;Common uses:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Restyle a selfie (anime, film, collage, doodle)&lt;/li&gt;
&lt;li&gt;Preview hairstyles on the same face&lt;/li&gt;
&lt;li&gt;Repair scratches and fading on old prints, then add natural color&lt;/li&gt;
&lt;li&gt;Turn a casual selfie into a clean studio-style headshot&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;You do not need to write long prompts. Ready-made “looks” are editable starting points; you press Generate only when the description looks right.&lt;/p&gt;

&lt;h2&gt;
  
  
  Step-by-step: generate on image2img.net
&lt;/h2&gt;

&lt;p&gt;The flow runs in the browser. No desktop app, and no account for the free daily images.&lt;/p&gt;

&lt;h3&gt;
  
  
  1. Open the site and upload a photo
&lt;/h3&gt;

&lt;p&gt;Go to &lt;a href="https://image2img.net" rel="noopener noreferrer"&gt;https://image2img.net&lt;/a&gt;. Drop a file, paste one, or upload from your phone camera roll. Supported formats are JPG, PNG, WebP, and HEIC, up to 20 MB.&lt;/p&gt;

&lt;h3&gt;
  
  
  2. Pick a look — or type what you want
&lt;/h3&gt;

&lt;p&gt;The homepage lists ready-made looks such as Anime, Doodle in real life, Paper collage, 90s film, Old photo restored, Studio headshot, Travel pin, and Try 4 hairstyles. Each look is a ready-made prompt you can still edit. Choosing a look does not use a free image; nothing runs until you press &lt;strong&gt;Generate&lt;/strong&gt;.&lt;/p&gt;

&lt;p&gt;If none of the looks fit, type a short instruction in plain language, for example: “make it a watercolor” or “warmer light, soft grey backdrop.”&lt;/p&gt;

&lt;h3&gt;
  
  
  3. Generate, then keep editing if needed
&lt;/h3&gt;

&lt;p&gt;Press Generate and wait a few seconds. Happy with the result? Download it — full size, no watermark. Not quite right? Describe the next change in words. Each edit becomes a new version (V1, V2, V3…), and you can jump back to any of them while you work.&lt;/p&gt;

&lt;h3&gt;
  
  
  4. Optional: log in only to save
&lt;/h3&gt;

&lt;p&gt;You do not need an account for the five free daily images. Logging in only saves work to &lt;strong&gt;My Creations&lt;/strong&gt; so you can come back later. Results without an account are deleted after 2 days; saved results after 30 days.&lt;/p&gt;

&lt;h2&gt;
  
  
  Prompt tips that actually help
&lt;/h2&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Start from a look, then tweak one thing.&lt;/strong&gt; Pick Anime or 90s film, then change one phrase (“warmer flash,” “softer line work”) instead of rewriting everything.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Say what to keep.&lt;/strong&gt; Image-to-image already preserves face and pose; saying “keep the same person and pose, only change the background to a soft grey studio” reduces surprises.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Iterate instead of restarting.&lt;/strong&gt; If the hair is right but the lighting is off, edit the result with one new sentence. You build on the version you liked.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Match the job to the look.&lt;/strong&gt; Old scratched prints → Old photo restored. Profile picture → Studio headshot. Playful social post → Doodle or Paper collage. Landscape souvenir vibe → Travel pin.&lt;/li&gt;
&lt;/ol&gt;

&lt;h2&gt;
  
  
  Free vs paid (exact limits from the site)
&lt;/h2&gt;

&lt;p&gt;Figures below come from the live homepage and &lt;a href="https://image2img.net/pricing" rel="noopener noreferrer"&gt;pricing page&lt;/a&gt; as of late October 2026.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Free (no account needed)&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;5 images per day at 1K standard quality&lt;/li&gt;
&lt;li&gt;Resets at 00:00 UTC&lt;/li&gt;
&lt;li&gt;No watermark, full-size downloads&lt;/li&gt;
&lt;li&gt;Commercial use allowed if you own the rights to the upload&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;strong&gt;When you need more&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Daily free images are always used first; credits only start after that&lt;/li&gt;
&lt;li&gt;Standard 1K image = 1 credit; HD 2K (long side about 2048 px) = 2 credits&lt;/li&gt;
&lt;li&gt;Failed or blocked generations are not charged&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;strong&gt;Subscriptions (monthly)&lt;/strong&gt;&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Basic: $9.90 / month, 150 credits&lt;/li&gt;
&lt;li&gt;Pro: $19.90 / month, 400 credits&lt;/li&gt;
&lt;li&gt;Yearly billing saves 20% (Basic about $7.92 / month billed yearly; Pro about $15.92 / month)&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;&lt;strong&gt;One-time credit packs&lt;/strong&gt; (valid 12 months)&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;50 credits for $4.99&lt;/li&gt;
&lt;li&gt;200 credits for $14.99&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Refunds: full refund within 7 days if you used less than 10% of that purchase’s credits. Cancel anytime from Billing; packs have nothing to cancel.&lt;/p&gt;

&lt;p&gt;Content policy blocks NSFW, nudity, images of minors, real-person face swaps, deceptive deepfakes, and graphic violence. Blocked attempts do not cost a free image or a credit.&lt;/p&gt;

&lt;h2&gt;
  
  
  Who this is for
&lt;/h2&gt;

&lt;p&gt;People who want a new look for a selfie or profile picture; anyone repairing or colorizing old family photos; creators who need quick variations with every version one click away.&lt;/p&gt;

&lt;p&gt;If that matches what you need, open &lt;a href="https://image2img.net" rel="noopener noreferrer"&gt;image2img.net&lt;/a&gt;, drop in one photo, pick a look, and press Generate. Five free images reset daily at 00:00 UTC — enough to try a few styles before deciding on credits.&lt;/p&gt;




&lt;p&gt;&lt;em&gt;Written by Jeson. Tool referenced: &lt;a href="https://image2img.net" rel="noopener noreferrer"&gt;Image2Img&lt;/a&gt; — free image-to-image AI in the browser.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>ai</category>
      <category>tutorial</category>
      <category>beginners</category>
      <category>productivity</category>
    </item>
  </channel>
</rss>
