hey all
I recently put a little money into my credit balance to play with OpenAI’s TTS models. The voice quality is super high! (it seems like one of them, Fable, is higher quality than the others - it takes noticeably longer to process and sounds a lot more humanlike) They’re not particularly well-documented, so I’d like any assistance I can get.
The API seems to randomly drop out completely or slow down - usually by the fourth or fifth API call in a given minute. I’m not sure if this is ratelimiting (the docs page told me they only kick in at around 50 rpm, this is closer to 4-5).
Sometimes, by the fourth or fifth API call, the API won’t respond at all 
unless I wait a short while. This is rare but happens every so often. Sometimes if it doesn’t respond, I can wait awhile and eventually get a response (usually 2-5 minutes).
Does anyone know why this happens/if I’m doing something wrong? Is there any way to speed up the API (optimize text content, etc) besides streaming it?
Thank you for any help 