Compare //
VoxtMini and Auralink vs Superwhisper and Wispr Flow
Four ways to dictate in any Mac app: two free and open source, two with subscriptions.
In short //
Pick Superwhisper or Wispr Flow if you dictate on Windows, iPhone or Android too, need more than English, Chinese and Japanese, or want a team plan with admin controls. Pick VoxtMini for free dictation on local models that works offline, or Auralink to use the speech and AI services you choose with your own keys, also free and open source.
Side by side //
| Compare | VoxtMini | Auralink | Superwhisper | Wispr Flow |
|---|---|---|---|---|
| Price | Free, open source (Apache-2.0) | Free, open source (AGPL-3.0); you pay the services you choose | Free tier; Pro $8.49 a month, $84.99 a year or $249.99 once | Free with a weekly word limit; Pro $15 per user per month, or $12 billed annually |
| Where speech becomes text | On your Mac by default (Qwen3-ASR, Whisper or Apple Speech); a remote service only for a feature you set | The remote services you choose, with your keys (Doubao, Alibaba Cloud, StepFun or any OpenAI-compatible endpoint) | Local Whisper models on the Mac, or cloud models | Always in the cloud |
| Works offline | Yes, with local models | No: it uses remote services | Yes, with local models | No |
| Devices | Mac (macOS 15 or later, Apple silicon) | Mac (macOS 15 or later, Apple silicon) | Mac, Windows, iOS and Android | Mac, Windows, iOS and Android |
| Languages | English, Chinese and Japanese | Those of the speech service you choose | 100+ languages and dialects; translates them to English | 100+ languages |
| Source code | On GitHub | On GitHub | Not published | Not published |
Where Superwhisper or Wispr Flow is the better choice
- You dictate on Windows, iPhone or Android as well as on a Mac.
- You speak a language other than English, Chinese or Japanese (both list 100+ languages).
- Your team needs shared dictionaries, admin controls, SSO or a signed BAA (Wispr Flow's team plans).
- You want meeting notes in the same app (Superwhisper Meetings on Mac, Wispr Flow Notetaker).
- You use an Intel Mac: Superwhisper suggests cloud models there; VoxtMini and Auralink need Apple silicon.
Where VoxtMini or Auralink is the better choice
- You want dictation that costs nothing and whose code you can read and build.
- Your audio should stay on the Mac by default: VoxtMini uses local models unless you pick a service for a feature.
- You'd rather pay the speech and AI services directly, with your own keys and no server of ours in between (Auralink).
- You need no account: neither app has a sign-in.
FAQ
Yes: VoxtMini is free under Apache-2.0 and runs speech and language models on your Mac by default, offline once they are downloaded. It covers English, Chinese and Japanese.
No. Wispr Flow's data controls page says transcription always happens in the cloud. VoxtMini and Superwhisper can transcribe offline with local models.
VoxtMini runs models on your Mac by default; Auralink uses remote services only, with your own keys, which keeps it small. Both are free, open source and type at your cursor in any app.
Sources
Facts about other products checked on November 17, 2026, from their own pages:
- Superwhisper (pricing and FAQ) superwhisper.com/
- Wispr Flow pricing wisprflow.ai/pricing
- Wispr Flow data controls wisprflow.ai/data-controls
- Wispr Flow privacy wisprflow.ai/privacy