Apple’s built-in transcriber blows away Whisper
John Voorhees of MacStories took Apple’s new Speech framework, available to all developers, for a spin in the macOS 26 beta and got great results in making audio transcripts:
It’s still early days for these technologies, but I’m here to tell you that their speed alone is a game changer for anyone who uses voice transcription to create text from lectures, podcasts, YouTube videos, and more. That’s something I do multiple times every week for AppStories, NPC, and Unwind, generating transcripts that I upload to YouTube because the site’s built-in transcription isn’t very good.
I’ve been using OpenAI’s open-source Whisper system (mosty whisper.cpp) for a couple of years, and while it seems to be more accurate than Apple’s model, it’s also half the speed of the large-v3-turbo model I’ve defaulted to lately.
It’s great to see that Apple is in this game, and even better, that it’s handing the power of this model to app developers so they can built speech-to-text transcription features directly into their apps.