0.1.0 betamacOS 15+ · Apple silicon
Press fn in any app and talk. Auralink types what you meant, without the ums and false starts, and translates or rewrites by voice too.
3
features, one shortcut each
20+
language model providers
4
speech recognition services
0
accounts, analytics or crash reports
Features
Transcribe
fnTap fn, speak, tap again. The words are recognised, tidied by a language model and typed into the app in front: fillers and false starts gone, punctuation and numbers right, your own rules applied.
Translate
fn⇧Select text and press fn ⇧ to translate it into the language you choose. With nothing selected, Auralink translates what you say. The answer card switches to another language in place.
Rewrite
fn⌃Select a hasty note, press fn ⌃ and say "make this a polite email to my manager". Insert puts the answer over the selection; Continue asks for a change, as a conversation.
App enhancement
Group apps and websites and give each group a scene: "a team chat: short and casual", "email: complete sentences". Auralink matches the app or browser tab in front and can press Return for you after inserting.
Dictionary
Add the names, products and terms recognition gets wrong, with the ways it tends to mishear them. They go to the speech service as hints and to the language model as a glossary. Import and export as CSV or text.
Overlay
A small pill near the bottom of the screen shows the feature, a waveform and the words as you say them. It never takes the focus from the app you type into, and the pointer goes straight through it.
Sound and silence
Before audio leaves your Mac, Auralink filters out hum, lifts a quiet voice and cuts the silence, with plain signal processing. Uploads get shorter, and services billed by the minute cost less. A recording with no speech is not sent at all.
Model services
Choose your speech recognition (Doubao, Alibaba Cloud, StepFun or any OpenAI-compatible endpoint) and your language model (DeepSeek, Qwen, Kimi, GLM, OpenAI, Anthropic, Gemini, OpenRouter, Groq and more). Test each connection, and pick one per feature.
More screens
Privacy
Auralink runs no server of its own. Audio and text go only to the speech and language services you set up, with your own keys.
Requirements
Permissions
Microphone
To record while a shortcut is active, and only then.
Accessibility
To type results into other apps, read the selected text, use fn as a shortcut and see which app is in front. Without it, results are copied instead.
FAQ
Yes. Auralink connects to the services you choose with your own keys, so you pay them directly, usually a fraction of a cent per minute. You need at least one speech service; a language model adds clean-up, Translate and Rewrite.
No subscription. Or get every paid app in the bundle for $39.
The first beta. Everything on this page is built and tested; the app icon is still a placeholder.