How is Raycast dictation so good?????
Literally end of message.
I've tried wisprflow, superwhisper, etc etc, local models, cloud models, windows, macos, etc. You name it, I’ve tried it.
I'm genuinely curious about HOW raycast is doing this. Like V2 is neat and all but this dictation is litearlly going to make me buy pro for sure. If it 8/month just for dictation without the rest of the app I might be tempted as well.
its SO fast. I do hundreds of words at once (daily journalling, writing emails, etc), and the amonut of time it takes to think makes it seem like they have to be processing the audio data as i'm speaking, in some sort of constant process loop or something. I'm not to familiar with the exact terminology, but I'm pretty sure the reason why other models take longer to think is because the audio has to be recorded, sent to the cloud, processed, refined (more processing), and then sent back to the user.
When you try it you'll understand. Anyone hav any ideas about how they were able to pull this off?