Monday, October 26, 2026 · Edition updated 6:00 AM ETMON, OCT 26 editionNext edition in Next in --:--:--
Intelligence Verdict
AI tools · Voice

OpenAI speech models

“The Audio API provides a speech endpoint based on our GPT-4o mini TTS (text-to-speech) model.” OpenAI, in its own words

  • Yesfree plan
  • $20a month, cheapest paid plan
  • 2places it runs: web, phone app
The facts

OpenAI speech models at a glance.

From OpenAI's own pages, read Oct 20, 2026. Prices can change at any time; check the maker's page before you pay.

Made by
OpenAIBased in US
What it is for
VoiceText read aloud in natural voices, and copies of a voice.
Free plan
YesChatGPT Free: 'Limited voice chats'; API is paid per token
Runs on
API, web, iOS, Android
Plans and prices
PlanA monthWhat it includes
API gpt-realtime-2.1 (audio)Not in US$$32.00 per 1M audio input tokens; $64.00 per 1M audio output tokens
API gpt-realtime-2.1-mini (audio)Not in US$$10.00 input / $20.00 output per 1M audio tokens
ChatGPT Plus$20Voice 'Expanded'
ChatGPT Pro$200Voice 'Unlimited*' subject to abuse guardrails

OpenAI API pricing, Oct 20, 2026 · company-reported

Read the small print

What its terms say.

Copied word for word from the maker's terms, with any lawsuits or complaints and their status when we last checked.

Who owns what you make
“As between you and OpenAI, and to the extent permitted by applicable law, you (a) retain your ownership rights in Input and (b) own the Output.”
OpenAI Terms of Use (effective 2026-01-01), Jan 1, 2026

Our note: TTS guide lists 13 built-in voices; custom voices require 'a speaker's consent recording and matching audio sample'. Usage policies require telling end users the voice is AI-generated. TTS-model per-minute prices were not shown on the pricing page fetched.