Javascript - compare audio tracks and check if they are the same word/phrase

I know there are great recognition APIs,plugin etc.

Just for my pourpose i would like to understand if is possible to compare 2 audio tracks and check if these 2 audio tracks contains the same/quite the same words/text inside.

So for example:

audio1.wav -> contains "Hey" spoke by google translate for example audio2.wav -> contains "Hey" spoke by a human

Is it possible to check if they matches at same words/senteces without using speech recognition ?