YOU ARE VIEWING ONE ITEM FROM THE AICRIER FEED

Hex 3.3.0 adds smarter local dictation

AICrier tracks AI developer news across Product Hunt, GitHub, Hacker News, YouTube, X, arXiv, and more. This page keeps the article you opened front and center while giving you a path into the live feed.

// WHAT AICRIER DOES

7+

TRACKED FEEDS

24/7

SCRAPED FEED

Short summaries, external links, screenshots, relevance scoring, tags, and featured picks for AI builders.

Hex 3.3.0 adds smarter local dictation
OPEN LINK ↗
// 1h agoPRODUCT UPDATE

Hex 3.3.0 adds smarter local dictation

Hex 3.3.0 improves onboarding, model selection, multilingual dictation, and desktop integration across macOS, Windows, and Linux. The release adds Russian GigaAM v3 support, traditional Chinese UI, per-mode vocabularies, Linux local API controls, and more resilient shortcut and paste handling.

// ANALYSIS

Hex is becoming a serious privacy-first alternative to cloud dictation by making local speech recognition easier to configure and more adaptable to real desktop workflows.

  • –The side-by-side model picker and recoverable setup flow reduce the biggest barrier to local dictation: choosing and downloading the right speech model.
  • –GigaAM v3 support and custom GigaAM files broaden language and model flexibility beyond the usual Whisper-centric ecosystem.
  • –Per-mode vocabularies, desktop-aware shortcut availability, and clearer paste-permission guidance make Hex feel increasingly tailored to professional workflows.
  • –Linux gaining local API controls brings feature parity closer across platforms and opens Hex to automation and developer integrations.
  • –The fixes for stuck microphones, injected shortcuts, clipboard preservation, and macOS key handling target reliability issues that can quickly undermine a voice-first tool.
// TAGS
hexspeechsttlocal-firstopen-sourceedge-ai

DISCOVERED

1h ago

2026-11-03

PUBLISHED

21h ago

2026-11-02

RELEVANCE

8/ 10