TOPLISTS / PROFILE INDEX
Tool / AI voices
OpenAI Speech
Text-to-speech in an existing OpenAI stack
Reviewed Nov 9, 2026 AI voicesEditorial take
Shortlist when your application already uses OpenAI APIs and you want a speech endpoint.
What to check
Follow the provider’s synthetic-voice disclosure requirements and test pronunciation.
Key facts
- Category
- AI voices
- Reviewed
- Nov 9, 2026
- Website
- developers.openai.com
Appears in
Check before you buy
- Price
- Usage-based. Realtime audio input and output have separate prices.
- Billing unit
- Estimate the full session with the current Realtime price table, including audio and text units.
- Key strength
- A Realtime API path for audio input and output with tools in the same conversation.
- Integration
- Realtime API. The older Speech API uses a different workflow.
- Data handling
- Consult the data-controls table for the exact endpoint and feature. Retention and zero-retention eligibility depend on the setup.
- Commercial use
- Review the business terms and the voice disclosure instructions before you deploy.
- Pilot check
- Prototype on the recommended replacement before committing. Check disclosure, interruption, tool calls and the cost of a complete conversation.
What changed
The official migration guide points new integrations to gpt-realtime-2.1-mini through the Realtime API. Recheck the deprecation notice before scheduling a migration.
Research Toplists editorial desk Reviewed November 9, 2026 Public documentation review