What are the three types of shadowing?

There's no official taxonomy of shadowing, but when people say "three types" they usually mean how much support you keep while you do it: shadowing blind by ear with no text, shadowing along with the transcript in front of you, and reading the text with a translation side by side first so you know what you're saying.

• Blind shadowing. Just the audio and your voice, no text. The hardest version and the best ear training, once you can nearly follow the clip.
• Shadowing with the text. You read along while you echo, so you can see the words you're chasing. The usual place to start.
• Side-by-side first. You read the native text next to a translation before shadowing, so meaning is locked in and you're only drilling the sound.

You'll also see other splits in language-teaching writing, for example complete versus selective shadowing (echoing everything versus catching only key phrases), sometimes labelled full, slash and silent shadowing. Nobody owns the definitions, so don't get hung up on the labels. The practical move is the same: start with the text visible and take it away as the clip gets comfortable, which is exactly how to shadow properly.