Voice Mode

AI · Segmented, VoiceMode

Talking to an agent instead of typing: same agent, same tools, same approvals, same thread.

Props: name, tint, state ("listening" | "thinking" | "speaking" | "idle"), stateLabel, transcript, muted, onMute, onEnd.

Approvals asked by voice still show the ApprovalCard; money needs the passkey on screen.

Props

Prop	Type	Notes
`options`	`(string \| Option)[]`	Required
`value?`	`string`
`defaultValue?`	`string`
`onChange?`	`(v: string) => void`
`size?`	`"sm"`
`block?`	`boolean`
`label?`	`string`
`className?`	`string`	Extra class on the root element.
`style?`	`CSSProperties`	Inline style on the root element.

Prop	Type	Notes
`name`	`string`	Required
`tint?`	`Tint`
`state`	`"listening" \| "thinking" \| "speaking" \| "idle"`	Required
`stateLabel?`	`string`
`transcript?`	`ReactNode`
`muted?`	`boolean`
`onMute?`	`() => void`
`onEnd?`	`() => void`
`className?`	`string`	Extra class on the root element.
`style?`	`CSSProperties`	Inline style on the root element.