I am trying to use the text-speach API to generate the pronunciation of one word in different languages. I have been trying to follow https://platform.openai.com/docs/guides/text-to-speech but I could not find a way to specify the language and that makes it impossible to use for this as words can sound very different in each language. The word “auto” is pronounced differently in spanish, german and english. Any ideas? Of course I could prepend a bit of text in each language to guide the model, but then I would have to know where exactly to clip the sound and it just becomes complex.
1 Like
I saw that Mac app has a main language selector. It could be very useful to extent language as a parameter for api call /audio/speech