It seems Apple has a number of under-the-hood AI enhancements within the works for iOS 26 and macOS Tahoe. Whereas a lot of the options are constructing on what’s already accessible, the corporate may even provide a chatbot-like expertise for individuals who’d like to speak to Apple Intelligence privately by the Shortcuts app, and it has an excellent speech API that outpaces OpenAI’s Whisper.
At the least, that’s what MacStories‘ John Voorhees claims in his hands-on report. He requested his son to construct Yap, a “easy command-line utility that takes audio and video recordsdata as enter and outputs SRT- and TXT-formatted transcripts.”
In his checks, he was capable of transcribe a 7GB 4K video model of a 34-minute-long AppStories podcast episode in solely 45 seconds and generate an SRT file. After doing the identical with different AI transcription fashions, Apple’s outperformed all of them:
- Yap: 45 seconds.
- MacWhisper (Massive V3 Turbo): 1 minute and 41 seconds.
- VidCap: 1 minute and 55 seconds.
- MacWhisper (Massive V2) 3 minutes and 55 seconds.
Whereas Apple’s AI transcription mannequin isn’t flawless, and it nonetheless had bother with final names and phrases like “AppStories,” Voorhees was impressed by Yap’s pace, being 55% sooner than OpenAI’s greatest mannequin whereas reaching the identical transcription high quality.
That stated, as soon as iOS 26 and macOS Tahoe are launched, you’ll in all probability see new apps benefiting from Apple’s newest AI fashions to investigate speech and transcribe information. Since these fashions are free for builders to make use of, they are going to enhance the marketplace for audio transcription.
Presently, these options are restricted to builders working the beta variations of iOS 26, macOS Tahoe, and Xcode 26.
