How do I use Happyly's voice modes?

Happyly has three voice modes, each suited to a different moment. None of them are hold-to-speak — you tap once to start, and Happyly handles the rest.

1. Microphone (record-and-edit) — free.

Tap the microphone icon next to the chat input. A waveform appears in place of the input — keep talking until you're done, then tap ✓ to confirm or ✗ to cancel. The transcript drops into the chat input where you can edit it before tapping Send. Best when you want a chance to fix a mishearing before the message goes out.

This mode is read-only at first — Happyly replies in text, not aloud.

2. Voice button (hands-free conversation) — free, the default.

Tap the Voice button on the right of the chat input. The voice panel opens above the input. Just talk — Happyly listens, detects when you've stopped (about a second of quiet), transcribes, replies in text and aloud, then automatically listens for your next turn. Tap the status pill while Happyly is talking to interrupt. Tap Stop to end the session.

A 10-minute hard cap closes the session if you forget. Best for everyday hands-busy moments — feeding, soothing, walking with a stroller.

3. Driving (real-time conversation) — Premium tier.

Flip the Driving switch inside the voice panel. The transport switches to a real-time channel: replies start within a second, and you can talk over Happyly to interrupt — no need to wait for it to finish. Same hands-free feel as the Voice button, just faster and more conversational.

Driving is the cost-heaviest surface (real-time audio is expensive), so it's gated to the Premium tier with a fair-use cap of 60 minutes a day, 1,500 minutes a month. The Microphone mode and the Voice button stay free regardless of tier.

Whisper Mode for the middle of the night.

While a voice session is active, tap the gear in the voice panel and switch on Whisper Mode. It drops Happyly's volume by half, dims the screen, switches to the Willow voice, and trims replies to one sentence. Happyly suggests it once after 10pm so you don't have to remember.

Picking a voice.

My Account → Preferences → Voice. Five voices to choose from: Willow, Sunny, Cedar, Quartz, Sage. Pick by ear — they're all calm and none are jarring.

What's stored.

The original audio is never saved — across all three voice modes. Whatever you said gets transcribed in real time, the transcript goes into chat history exactly like a typed message, and the audio itself is gone. You can delete or export voice transcripts the same way you would typed messages.

What voice can do.

Log sleeps, feeds, pumps, and diapers. Discard an in-progress sleep, feed, or pump. Log growth measurements, milestones, medication doses (with a verbatim readback before logging — "Just to confirm: 2.5 ml of paracetamol for Arlo at 8pm — say yes to log it."), and vaccines. Switch active child. Change theme. Play and stop sleep sounds, set a sound timer. Open a guide. Hand off on duty ("I'm on now" / "You take over").

What voice can't do yet.

Edits or deletes of past entries (typed precision wins for these). Adding, editing, or removing guests (privacy boundary — has to be tap-confirmed). Deleting a child profile (catastrophic by voice). Calendar navigation. Profile, notification, or unit settings. Sharing or revoking the Rhythm Guide. Milk Stash triage beyond simple add / use. If you ask Happyly to do one of these by voice, it'll say so honestly and point you to the typed-chat path.