Add audio input

[Laged]

It should send it to an endpoind in the backend that's TBA for now :D

[Laged]

https://docs.api.ai/docs/query#post-query-multipart

"Takes natural language text as a sound file and returns information as JSON.
Currently, the sound files must be 16000 Hz, Signed PCM, 16 bit, and mono."

WAV type=audio/wav

Raw audio (headerless) type=audio/x-raw