The only format we recommend using is MP3. MP3 by default when downloading or rendering has good file sizes and audio quality for listening. These days essentially every audio you hear across the internet or device is typically in MP3 format, even in videos.
Bitrate is similar to the data rate in videos. It is how much data is being stored for the audio every second it is recording. Most audio files and videos have an audio bitrate of 90-130kpbs, this size is perfectly fine for listening.
The sample rate is the quality of the audio is played at, similar to the data rate for videos. It is recommended to keep it at around 48kHz or higher, depending on if you have professionally recorded the audio or if it is just a general audio file.
Don't go too high as it won't affect or go over the sample rate of what is recorded.
This affects how many speakers or audio outputs the audio will be played. These days the default is 2 (Stereo) and works for everything. If you are to use Mono, ensure that you only you 1 speaker/audio device output, but we still recommend keeping this to 2.
Typically, audio files out of most files are sometimes the smallest, but we recommend not inflating the files by making rendering/recording settings over the amount needed, doing this will not magically increase the quality of the audio and will just result in an unneeded large file size.
While this isn't necessarily an important subject, in order to keep workflow going well and ensuring you know what each file is, we recommend naming the audio something you will know when you first glance at it.
Giving your audio a randomly generated name can slow down work time for yourself when building your publication, but when you know the file name in correspondence to the audio you will know straight away and won't have to waste time listening to the audio to know what it is.