10  Accessibility & Speech

NeoWeatherwall is built to be usable by everyone and to keep you informed hands-free during active weather. This chapter covers text-to-speech and the app’s accessibility support.

10.1 Text-to-speech (TTS)

The app can read NWS products (warnings, watches, and MCDs) aloud.

10.1.1 Voice tiers

Choose a voice tier under Settings > Speech:

  • Default: Apple on-device premium/neural voices. Fast, free, works offline.
  • Enhanced: higher-quality on-device Apple voices (one-time download).
  • Premium (opt-in): cloud voices (ElevenLabs Turbo v2.5 or OpenAI TTS) for the most natural speech. Requires an API key and a network connection.

Within a tier, pick the specific voice.

Screenshot of the speech settings with voice, rate, and pitch controls

Speech settings: voice tier, voice, rate, and pitch.

10.1.2 Playback controls

  • Rate: adjustable speaking speed (0.35–0.65).
  • Pitch: adjustable voice pitch (0.8–1.2).
  • Preview: play a sample phrase with your current settings.
  • Mini player: appears during playback with play/pause/stop and progress.

Screenshot of the speech mini player overlay

The mini player overlay shown during speech playback.

10.1.3 Reading products aloud

  • Manual: tap the speaker button on a warning, watch, or MCD detail sheet to hear it.
  • Auto-Announce: optionally announce new critical products (Tornado, Severe Thunderstorm, and Flash Flood Warnings) the moment they’re issued.
NoteSmart announcements

Auto-announce is throttled so you don’t hear duplicates. It won’t repeat the same product, and it limits announcements to roughly one every 30 seconds. NWS jargon is normalized for natural speech, so “TSTM” is read as “thunderstorm” and 1842Z as “eighteen forty-two Zulu”.

10.2 VoiceOver and system accessibility

  • VoiceOver: all controls carry proper labels, hints, and identifiers. When VoiceOver is active, auto-announce steps aside so you don’t hear things twice.
  • Reduce Motion: when the system setting is on, animations are disabled or simplified.
  • Dynamic Type: text responds to your system/app font-size preference.
  • High-contrast overlays: enhanced contrast options for UI elements.
  • Keyboard navigation: full keyboard support on macOS.

10.3 Audio behavior

Speech uses a spoken-audio session that plays well with other audio and works on the Lock Screen and CarPlay. Non-critical speech respects the silent switch. Critical warnings are designed to play even when the switch is on.