GPT-4o mini TTS 🧠 Model Paid

What it is

OpenAI's text-to-speech family is API-first. The current recommended model is gpt-4o-mini-tts. Its latest snapshot is dated December 2025. You tell it what to say and also how to say it: tone, emotion, pace and accent come from a plain-language instruction. The older tts-1 and tts-1-hd are still available. For live voice agents, OpenAI points to its separate Realtime models (gpt-realtime).

Key features

• 13 built-in voices. OpenAI recommends marin and cedar for best quality
• Style instructions in plain language, with no SSML needed
• Follows Whisper's language coverage, but voices are optimised for English
• Streaming output for low-latency playback
• Custom voices for eligible customers, with a recorded consent statement from the speaker

Where you can use it

Mostly developers and automation tools calling the OpenAI API directly. We found no creator app in our directory that documents gpt-4o-mini-tts by name. ChatGPT's own voice features run on OpenAI's realtime voice stack instead.

Pricing and rights

Usage-based pricing only: $0.60 per million input text tokens and $12 per million audio output tokens. There is no subscription. OpenAI's usage policies require you to tell listeners clearly that the voice is AI-generated. Custom voices need the speaker's consent recording.

Who it is for

Developers building narration, accessibility or app voice features who already use the OpenAI API and want cheap, steerable voices.

Verdict

GPT-4o mini TTS is inexpensive and easy to steer, and it is a sensible default inside OpenAI-based pipelines. The voice list is small, non-English delivery trails ElevenLabs and Gemini, and there is no consumer editor. It is a building block, not a creator app.