In a SPEECH session, your audio is converted into text. You can determine where the audio signal comes from in the session via the “Ingest” tab (input source).
By default, you send the signal directly from the browser in which you have just opened the session by clicking on “Start audio transmission” in the top right corner and selecting the audio source.
It is also possible to control the Speech Session (and audio transmission) via our API interfaces. You can find out more about this here.
You can also send the audio signal from a livestream encoder. To do this, you need to enter the RTMP URL and stream key in your encoder. You can find these in the “Ingest” section under “RTMP” – from there, you can simply copy and paste them.
Once you have entered the data into your encoder (below you can see what it looks like in the virtual encoder “OBS”), start the stream in your encoder and the audio signal will arrive in the SPEECH Session and be transcribed.