Hey there,
for the first time ever I am suddenly seeing lots and lots of timeout exceptions when generating TTS via the api. I am using the model “gpt-4o-mini-tts-2025-12-15“.
Anyone else experiencing this? Generation of a short Sentence “Hey, my name is Mark and I am available” is taking 15-20 seconds.
Can’t be rate limited, I am at 0.62 cents today.
Something is up. It got super slow and now I’m not getting anything. Oddly, my usage says .62 also.
Hey @Dortmund_Dev and @sclayturner, this is definitely odd, and there were no reported outages on March 7th.
Can you let us know if you are still seeing the same behavior? If yes, is it with the same TTS endpoint and gpt-4o-mini-tts model?
Also curious if you have tried using any other models or other API endpoints and whether you are seeing the same timeouts and latency.
1 Like