I’m really trying to understand the logic behind the decision of giving builders access to a much less realistic, human-sounding and overall good version of realtime voices (possibly a different model altogether)?
i mean the differences between AVM in the chatgpt app and the realtime-api are night and day, and are literally making builders seeking other realistic solutions such as google’s native audio models.
this isn’t just a rant - as a builder that loves working with openai’s products and ecosystem, i (and i’m sure many builders out there) would like to know if we should expect the AVM model to be available for api usage? is it on the roadmap? any timelines whatsoever?
again, this is crucial information that will affect actual, real world product development.
Thank you in advance.