Add WAV Audio Speech Response Format

[nezhyborets]

Is your feature request related to a problem? Please describe.
As mentioned by @SunburstEnzo in this PR, OpenAI's text-to-speech supports wav output format. We should add that too.

Describe the solution you'd like
The PR above shows the solution, simply add case wav to the enum should work, but should be tried out if it actually works.