Hey! So I have something new to announce. It's still early, and kind of an experiment for me: a new iPhone, iPad and (native) Mac app called Transcriptionist. It takes speech and turns it into text — privately, on-device — and a bunch more besides.
It can be used as a companion app to Ferrite (a new update will integrate to transcribe your shows) but is a great stand-alone app too.
• It's accelerated with the Apple Neural Engine, so it's very fast
• It can transcribe almost 100 languages
Another experiment is that I've opened up around 100 slots on a public TestFlight. Haven't decided whether I'll keep it open, but I've not tried it before and I wanted to give it a shot. So if you would find this useful: https://testflight.apple.com/join/ppYPASJO
@wooji looks awesome, can’t wait to try it! will it have speaker detection for a single file with multiple speakers in it? (ie two+ people, one mic)
@daniel unfortunately not, it’s a feature I’d love to have, but don’t know of a good way to do that on-device
• Multitrack transcription: put each person on their own track in Ferrite (or record them to their own file & import all into Transcriptionist) and it'll show who said what. Also means it doesn't get thrown off by crosstalk.
• Of course, can play back audio (up to 2x speed) & tracks progress alongside speakers.
• Search (inc. regexps) and replace.
• Export in a bunch of formats — or create your own with built-in editor.
• Wide accessibility support: VoiceOver, Voice Control, Dynamic Type etc