Question 1

Where can I dictate?

Accepted Answer

Use NeverType in supported text fields in apps such as Slack, Google Docs, email, and ChatGPT. Automatic insertion depends on the app, operating system, and permissions.

Question 2

What is Instinct?

Accepted Answer

Instinct is optional English cleanup after dictation. It can fix grammar, trim repetition, apply spoken corrections, and use available app context to format messages or structure AI prompts. Choose a local model or bring an API key for a supported cloud provider. Your original transcript stays in History.

Question 3

Does everything stay on my device?

Accepted Answer

When you choose local speech and local Instinct models, inference runs on your computer. Initial model downloads require internet access. If you enable a cloud engine, the relevant audio or transcript and enabled context are sent to your selected provider. Cloud API usage is billed separately by that provider.

Question 4

Which computers can I use?

Accepted Answer

NeverType is available for macOS on Apple Silicon and Intel, and for Ubuntu / Linux. There is no Windows version. Choose the installer for your computer in the download section.

Question 5

What languages can I dictate in?

Accepted Answer

Language support depends on the speech model you choose. The model manager shows each model’s languages and whether it supports live transcription. Instinct currently formats English only; other languages keep the original transcript.

Question 6

Does it know which app or website I’m using?

Accepted Answer

Instinct can use the foreground app and available website or field hints. On macOS, Accessibility access can provide a short excerpt around your cursor. Coverage varies by app and operating system. Unknown destinations use general cleanup, and you can turn app context off.

Question 7

Why is there a wait after I stop speaking?

Accepted Answer

Speech recognition finalizes your transcript first. If Instinct is enabled, it then runs a separate formatting pass. Cleanup can add a short wait. You can skip it while it runs or set a maximum wait. If cleanup fails or times out, NeverType uses the original transcript.

Question 8

Are speech models and Instinct models the same?

Accepted Answer

No. Speech models turn audio into text. Instinct text models help rewrite and format that transcript. Each has its own selection in the app. For cloud transcription, you can use Groq, OpenAI or Deepgram. Instinct offers additional text-model providers. Your provider bills API usage separately.

Release History

v1.0.11

v1.0.10

v1.0.9

v1.0.7

v1.0.6

v1.0.5

v1.0.0