First Run and Onboarding

Onboarding Overview

Steno onboarding has four steps:

1. Welcome
2. Permissions
3. Local transcription setup
4. Feature tour

The flow is designed so you can start quickly while still understanding the important controls.

Permissions Step

Steno checks three permission states in-app:

• Microphone
• Accessibility
• Input Monitoring

Important behavior:

• microphone permission is required to continue onboarding
• accessibility and input monitoring can be skipped, but key features will be limited
• when the app becomes active again, permission states are refreshed

Engine Step

You provide or confirm two paths:

• whisper-cli binary path
• model file path (for example ggml-small.en.bin)

Onboarding validates both paths by checking file existence. You cannot continue past this step until both are valid.

For better silence and background noise suppression, the onboarding step also recommends downloading the optional VAD model (Silero). VAD is enabled by default when a model is available.

Finish and Verify

At the end of onboarding, Steno saves:

• updated whisper/model paths (if changed)
• onboarding completion flag

Quick verification after clicking Get Started:

1. hold Option and dictate a sentence
2. test hands-free toggle
3. confirm a transcript appears in History