Can I use OS X 10.8's speech recognition/dictation without a GUI?

Question

Ideally I would like to use 10.8's/Mountain Lion's dictation feature from Python, without a GUI. As far as I can tell dictation is currently only supported by using the UITextInput protocol, which implies a GUI (correct?).

Vedic

Is there any particular reason why you tagged this under python? Are you looking to with it in python? – Zabaglione

Yes, as I said in my question, ideally I would like to use the dictation feature from within Python. – Vedic

Oh my bad I admit to skimming... – Zabaglione

You might also want to follow this related question: #11703304 – Rigorism

I'm pretty sure UITextInput is how you use iOS's dictation feature from iOS, and has nothing to do with OS X, so you've been searching in the wrong place. – Brahmana

June 2015 -- Bump! Did you ever get anywhere with this? – Fewer

Sorry to say I did not. – Vedic

Answer 1

You can create a dummy text field and programmatically start dictating by emulating the keypress (tried out successfully here)

Rigorism

@Rigorism Can you give me a hint how to emulate the "fn" key press twice? – Thulium

CGEventCreateKeyboardEvent(src, 0x3F, true) it is :) – Thulium

Answer 2

I think you want the Speech Framework. SFSpeechRecognizer et al. I don't believe this requires a window session on macOS as this is part of AVFoundation, though I'm not 100% certain. Try it and see. :-)

Grey