feat: mute-to-compose mode (mic off, text input, TTS playback) #12

New issue

Open

opened 2026-04-06 13:11:41 -07:00 by pyr0ball · 0 comments

pyr0ball commented

2026-04-06 13:11:41 -07:00

Owner

Allow users who cannot produce speech in real time to compose text responses that are read aloud by TTS.

Requirements:

Mic mute button with clear visual state (large, accessible, thumb-zone placement)
When muted: text compose field appears (keyboard-first focus)
Compose latency communicated (spinner while TTS queues — never silent loading)
TTS backend: local (Free tier) or cloud (Paid tier)
Auto-send timeout: configurable (default: none — user must confirm send)
Compose window dismissable without sending
Session annotation continues passively while muted (listening to other speakers)

ND/adaptive constraint: this mode exists for AAC users, selective mutism, and processing-delay users. No urgency framing. No 'recording paused' red states.

Allow users who cannot produce speech in real time to compose text responses that are read aloud by TTS. Requirements: - Mic mute button with clear visual state (large, accessible, thumb-zone placement) - When muted: text compose field appears (keyboard-first focus) - Compose latency communicated (spinner while TTS queues — never silent loading) - TTS backend: local (Free tier) or cloud (Paid tier) - Auto-send timeout: configurable (default: none — user must confirm send) - Compose window dismissable without sending - Session annotation continues passively while muted (listening to other speakers) ND/adaptive constraint: this mode exists for AAC users, selective mutism, and processing-delay users. No urgency framing. No 'recording paused' red states.