Video chat android app with transcription (using Nuance Transcription Services). Two users can connect and talk. One user can see the other user's face in full screen, but there is a small area (bottom right) where the user sees himself/herself. When one user presses the phone screen and speaks, the app records the voice and sends it to Nuance's API, which returns the transcription text (displayed on the screen of the other user). The client / server communication runs over the UDP protocol.

Who do I talk to?