Best practices for audio file to speech conversion

Can you folks offer some best practices for audio file to speech conversion using TextEdit.

I have downloaded various clips from YouTube, individuals doing speeches at conferences, that I want to translate to text so I can identify which clips discuss certain topics.

These clips are sometimes long ... for this discussion lets say longer than one minute.

The clips are tossed into Audacity as there I can choose Soundflower as the output and in the Sound control panel I select Soundflower as the input.

(Soundflower lets you channel audio to the a sound card instead of the speaker. (It allows different applications to access soundcard at the same time. It's a critical utility for a music producer for example.)

The method is you start the audio in Audacity, open TextEdit, Start Dictation and you can see the text being translated and inserted into the TextEdit document.

(You can visually see the mic icon move ... one can't hear the audio, it's being passed internally.)

But it seems the Mac can't keep up because after 5-10 seconds of speech, just one person, the text stops being inserted into the TextEdit window.

There's nothing indecipherable about the audio ... if I stop and start the dictation again the next 10-15 seconds are picked up.

And this is not a weak Mac ... it's a 2017 MacBook Pro; 16gb; 3.1G

So... finally... the question is ... is there something I can do to the audio that will make it easier for the Mac to keep up. Speed, pitch, removing any echo?

Yes, I could play with dozens of different settings in Audacity but I'd simply rather not guess if you folks already have some expertise in this as I am not a audio engineer.

Thank you. Randy