TypeWhisper 1.4 release candidate adds unified workflows, system-wide dictation, and support for ten transcription engines on macOS

TypeWhisper 1.4 is the current release-candidate line for macOS, featuring system-wide dictation, file transcription, unified workflows, history, dictionary, snippets, and bundled integrations. Advanced surfaces like HTTP API, CLI, widgets, watch folders, and plugin SDK remain supported.

Transcription can run on-device using WhisperKit (99+ languages, streaming, translation), Parakeet TDT v3 (25 European languages), Apple SpeechAnalyzer (macOS 26+), Granite Speech, Qwen3 ASR, Voxtral, or via cloud APIs from Groq, OpenAI, xAI/Grok, and OpenAI-compatible providers. Workflows enable reusable transformations (translation, rewriting, extraction, formatting) triggered per-app, per-website, by hotkey, or globally.

New in 1.3: Unified Workflows consolidate prompt actions and matching rules in a native editor; Always Fallback Trigger runs a global workflow when no specific app or website workflow matches; Manual Workflow Palette keeps workflows out of automatic dictation and runs them from a global shortcut; Safer Prompt Boundaries treat dictated text as source content to transform, not instructions; Focus-Safe Local Processing preserves focus in the original target app during on-device workflows.