AKHY AHMEDKABIR
1.3K posts
@kabslock123
« Ô Mes serviteurs qui avez commis des excès à votre propre détriment, ne désespérez pas de la miséricorde d’Allah. Car Allah pardonne tous les péchés »
Imagines nga nekk ci sa bàmmeel yaw rekk, dóotoo gis ñi nga daan gis, dóotóo lekk li nga daan lekk, dóotóo dem foo bëgg dem…nekk ci benn barab rekk di xaar ñu àtte la, xamóo ci sa bopp dara, te wóorul sax ñaan fekk la fa nga nekk ndax waa àdduna duñu gaaw a fàttali ku faatu…
you don't need to pay for ElevenLabs & WisprFlow subscription anymore. someone just open-sourced BOTH in one app. completely free. runs locally. no API keys. no usage caps. → clone any voice → global dictation hotkey → your agent can talk back → 7 TTS engines → 23 languages here's how to set it up under 5mins: 1/ download mac (apple silicon): voicebox.sh/download/mac-a… mac (intel): voicebox.sh/download/mac-i… windows: voicebox.sh/download/windo… docker: docker compose up open the dmg, drag to applications, launch. 2/ clone your voice (60 sec) → profiles tab → + new profile → name it, pick a language → upload a 10–30s clean audio sample (or record live in-app) → save that’s your voice. reusable across every engine. 3/ generate speech → generate tab → pick your profile → type your text → hit generate first run downloads the model (one-time). after that it’s a few seconds per clip. pro tip: use chatterbox turbo + type / in the box to drop in [laugh] [sigh] [gasp] tags. 4/ give your AI agent your voice this is the part that broke me. voicebox ships with an mcp server. add it to claude code / cursor / cline once, and your agent calls voicebox.speak to talk back in your cloned voice. settings → mcp → copy the snippet into your agent’s config. done. 5/ dictate into anything → settings → dictation → set a global hotkey → hold it anywhere on your system, speak, release → transcript pastes into the focused text field slack, email, code, browser — works everywhere. 6/ that’s it. no api keys. no usage caps. no audio leaving your machine. you just replaced two SaaS subscriptions with a 200mb native app. github.com/jamiepine/voic… follow and turn on post notification for more content that helps you maximize the use of AI.
Google Chrome is quietly downloading a roughly 4 GB AI model to many users’ computers without clear upfront consent. The file, called weights.bin, is part of Google’s Gemini Nano on-device language model and lands in the browser’s user data folder under OptGuideOnDeviceModel. It powers built-in AI tools such as “Help me write,” smarter tab suggestions, on-device scam detection, and page summarization. The download triggers automatically for devices meeting minimum hardware requirements, and Chrome often replaces the files if deleted. While the model processes data locally, installation happens in the background with minimal notification. The scale is noteworthy. Hundreds of millions or billions of installations add up to thousands of tonnes of carbon emissions globally from data transfer, even though each is a one-time event. To prevent or remove it, go to chrome://flags, disable the entries for the optimization guide on-device model and Prompt API, restart the browser, and manually delete the folder.
