New TTS model (gpt-4o-mini-tts) Ignoring Speed parameter

,

It seems you are incorrect. I tried the new model again recently and it is no longer ignoring the speed parameter. So apparently it is indeed supported?