OpenAI makes custom voice generation available to every GPT-Live API user
In short: OpenAI has rolled out prompt-to-voice access to all GPT-Live API customers, allowing a spoken voice to be designed simply by describing it in text, either through the API or the voice Playground, and then reused inside live voice sessions. Designing the voice costs nothing, but speaking with it still runs on normal GPT-Live session billing. Alongside this rollout, nine additional built-in voices covering German, French, Hindi, Japanese and Korean were added to the platform.
This summary was generated automatically by AI from OpenAI API Changelog's publication. It is our own text, not a copy of the original — facts, figures and quotes belong to the source, linked above and below.
What changed?
- 1Prompt-to-voice access expanded from a limited group to all API customers
- 2Voices can be designed via v1/audio/voices or the Playground, then used in v1/live/sessions
- 3Designing a voice is free; actual speech output still follows standard GPT-Live session pricing
- 4Nine new voices added: Nuri, Juni, Flitz (German), Brise, Sillage (French), Nira (Hindi), Harema, Shida (Japanese), Noeul (Korean)
| Parameter | Before | Now |
|---|---|---|
| Voice creation access | Limited availability | Available to all API customers |
| Voice creation cost | Not specified | Free |
| Speech generation billing | Standard GPT-Live session billing | Standard GPT-Live session billing (unchanged) |
| Built-in voices | Existing set | +9 new voices: German (Nuri, Juni, Flitz), French (Brise, Sillage), Hindi (Nira), Japanese (Harema, Shida), Korean (Noeu |
Why it matters
Teams building spoken agents now get a much larger pool of ready-made voices plus a low-friction way to design a bespoke voice without extra training or special access, speeding up localization and brand-voice work.
What it means for AI agents and contact centers
This matters for anyone running voice agents on GPT-Live, since free voice design and new non-English voices could reduce the cost of adding new languages or giving an agent a distinct identity; the new voices don't include Lithuanian, so impact on Lithuanian-language deployments is limited to the broader platform improvement.
🧪 Worth Testing
Testing would reveal whether the free custom-voice design produces consistent, production-ready quality and whether the new multilingual voices meet the bar for live contact-center use.
Sources
- OpenAI API ChangelogOfficialPrimary sourceOriginal article →„Released prompt-to-voice for all API customers. Create a custom voice from a text description through the API or voice Playground, then reuse it in GPT-Live sessions. Creating a prompt-generated voice“16 Oct 2026, 03:00
- Published by source
- 16 Oct 2026, 03:00
- Found by our system
- 16 Oct 2026, 20:11
- Summary generated
- 16 Oct 2026, 20:13
This article was written by AI from the original source. Facts, numbers and prices come from the source; missing values are marked “Not specified”. Legal notice, copyright and privacy