I’m thinking it would be. Right or wrong, I don’t know.
With how much the letters B, C, D, E, G sound alike, I’d be very worried about accuracy, considering the kinds of errors I see StT making on a regular basis in things like auto-generated Youtube captions. Using the Shift-Q popover is already very fast, as your hands don’t need to leave the keyboard and you can be completely precise and deterministic in your input.
True. I suppose you could reassign the sounds.
Speech can’t beat a moderately-competent typist for this sort of thing. My two cents.
2 Likes
Disagree there, though can’t prove it.
Speech can be recorded in the ram memory first and processed at whatever speed the interpreter goes. The ram will hold the recording until it finishes processing. Overall, should be faster than a 100 word per minute typist.
And if it can’t today, I suspect in a couple years or less it will be able to.
Also, for best results you’d need a stand-alone speech to text app, that doesn’t go through a massive process like google. Also voice training to your own voice should increase accuracy.
Also, a lot less language variation with just music related terms, than all of language.
The fastest way to input chords is certainly a midi keyboard. You can play chords right into the popover and it interprets them.
Yes. But it interprets them very idiosyncratically.
I find the interpretation pretty consistent, what little I’ve used it. You need to put the desired root at the bottom.
I didn’t suggest it was inconsistent.
Yes, I guess I should have said “fastest without a midi keyboard”
Some work for a while without a midi keyboard. Some work without even physically listening to the passage they are working on until they have finished it.