Voice Responses

Cortex can speak AI responses aloud using ElevenLabs text-to-speech. This is useful for hands-free review sessions, accessibility, or when you prefer listening to reading.

Voice responses are available on Student ($5/mo) and above. See Pricing Tiers for quotas.

Enabling Voice

In the AI Assistant, toggle the voice icon to enable spoken responses. The AI will both display text and read the response aloud.

Voice Models

Model	Latency	Quality	Cost
Flash	~200ms	Good	$0.08/1K chars
Turbo	~300ms	Better	$0.11/1K chars

Cortex uses Flash by default for speed. Turbo is used for longer, synthesized responses where quality matters more.

Usage Tracking

Voice responses count against your monthly quota:

Student: 15 voice responses/month
Pro: 60 voice responses/month
Power: 200 voice responses/month

Check your remaining quota on the dashboard or via the /api/quota endpoint.