| Speech path | Google says the current native Mac feature set runs 100% on-device and fully offline | Wispr says transcription always happens in its cloud; retention depends on Privacy Mode and Private Cloud Sync | Speech recognition and supported formatting run on the Mac after model setup |
|---|
| Platforms and Macs | Native macOS release; the reviewed Google pages do not publish complete desktop system requirements | Mac, Windows, iPhone, and Android; Wispr currently documents macOS 12+ on Apple-silicon or Intel Macs | macOS 26+ on Apple-silicon Macs only |
|---|
| Price and limits | Google promotes the Mac download as “No cap” and “without subscription,” but the reviewed Mac pages do not publish a permanent desktop price | Basic is free with 2,000 words per week on desktop; Pro is $15 monthly or $12 per month billed annually after a 14-day Pro trial | $50 once after the 14-day full trial |
|---|
| Languages | The current Google product FAQ says English only | Wispr advertises support for 100+ languages | Nine current English regional speech choices |
|---|
| Dictation and editing | Customizable hotkey dictation across Mac apps plus Voice Edit for selected text and spoken rewrite commands | Cross-app voice typing, custom dictionary and snippets; paid Pro adds Command Mode and unlimited words | Hold Right Option for Dictate or Globe for Voice-to-Spec, then review or insert the result |
|---|
| Saved media | Google documents fully local transcription of audio and video files | The reviewed pricing and privacy pages center on live voice typing rather than a comparable local saved-file workflow | Does not transcribe imported audio or video files |
|---|
| Privacy and retention | Google says the Mac feature set and user data stay on-device; optional Workspace or Gmail vocabulary access is a separate connected boundary | Cloud transcription; Wispr says zero retention requires Privacy Mode enabled and Private Cloud Sync disabled | No IraVoice server receives audio or dictated text; model download, one paid Gumroad check, and optional agent handoff are disclosed separately |
|---|
| Developer workflow | General cross-app dictation and Voice Edit; the reviewed sources do not document a guarded developer-spec handoff | General cross-platform voice typing and AI commands; the reviewed sources do not document a separate review-first spec guard | Voice-to-Spec creates reviewable Markdown and can use a guarded, visible Codex or Claude Code handoff |
|---|