Voice Input & Output

AI Council Chat keeps Speech-to-Text (input) and Text-to-Speech (output) cleanly separated. This split is intentional, both economically and operationally: dictation runs locally and free of charge, while playback consumes credits because it uses premium AI voices.

Dictation (STT) — 100% free

The microphone uses the native Web Speech API built into your browser. No server roundtrip, no credits, no quota. It works on desktop and mobile and supports many languages depending on your system.

Playback (TTS) — premium, but transparent

When you want the AI to read an answer aloud, premium neural voices are used. Billing is governed by three rules designed to protect you from any invoice surprises:

• Explicit click: TTS never auto-plays. You have to press the play button yourself.
• Price upfront: the exact cost (e.g., 1 credit / 50 characters) is shown next to the button before you confirm.
• Local cache: once audio is generated, it's stored in your browser. Pause, rewind, and replay all play from cache — you are never charged twice for the same passage.