TTS with a question... the audio back has the answer

You’re right that gpt-4o-mini-tts-2025-03-20 is listed for shutdown, with the current public deprecation date of July 23, 2026,: https://developers.openai.com/api/docs/deprecations.

For this specific /v1/audio/speech issue, the practical split is still: use tts-1 / tts-1-hd when strict verbatim readout matters most. Hope that helps. Thank you!