Scusi, from iOS and Mac app developer Jordi Bruin, lets you transcribe any audio message in the Mac’s iMessage app just by dragging it. This is similar to the transcription feature recently added to WhatsApp but might be of more interest to folks in the US because iMessage is much more popular there. Scusi makes use of the built-in accessibility tools in the Mac, showing how developers can easily add features to built in Mac apps—something almost impossible on the iPhone and iPad (although, if you’re iPhone-only, there is a kind of a workaround). “Apple’s Speech recognition APIs make it super simple to do any form of speech to text. The trickiest part is finding the audio file linked to the voice message, but once you’ve got that, it’s very easy to get reliable transcripts,” Scusi developer Jordi Bruin told Lifewire via email.
Never Ending Story
Voice messages are only getting more popular. If you haven’t yet experienced a message inbox full of messages that have to be listened to just to find out the subject, you’re one of the lucky ones. For the sender, it’s easy to see the appeal because you don’t have to type. You just press a button and talk. And because you’re talking, you probably aren’t nearly as focused as you would be if you had to type those words into a text box. For the person receiving the message, there are two possibilities. Either it’s a message from a loved one or a good friend, and you enjoy listening to them chit-chat and never get to the point. Or you hate it for the same reasons. “I hate receiving [voice messages] because often I can’t listen to them on the spot, and then I forget about them. Transcribing would be so nice,” wrote Scusi co-developer Hidde van der Ploeg on Twitter shortly before creating Scusi with Bruin. There’s probably nothing more efficient in terms of communication bandwidth than having an actual conversation with somebody. You can react, hash out problems, and do all the things we’ve been doing as a species for millennia. Instead of emailing or WhatsApping back and forth for days, jumping on a call for five minutes is way, way better. But voice messages are the opposite. They might be one of the least efficient ways to communicate. For example, let’s go back to the bad old days of leaving voicemails on people’s answering machines. Do you remember how people would only leave their phone number, or other essential detail, at the very end of a message? If you missed it, you had to listen to the whole thing again and hope you got it the second time around.
Usability
Bruin’s Scusi comes as a separate app for your Mac, but when you launch it, all you see is a new icon in the menu bar. But when you start to drag an audio clip from any conversation thread, a little window pops up for you to drop the clip into. Then, it transcribes it using the Mac’s built-in voice-to-text engine, which can do a good-to-great job depending on the quality of the audio clip. In testing, I’ve found this engine to be surprisingly good, and one advantage of the built-in engine is it all happens on your device, not on a server somewhere. Even if you like listening to never-ending voice messages, transcription is handy. It makes it easy to find that phone number or address or to preview a long message before listening, just to find out what it’s about. Unfortunately, right now, Scusi is Mac-only. You can get the app for free via Bruin’s Gumroad store page, although it may be coming to the Mac App Store in the future. “We started out with Gumroad because it allows for a very fast feedback loop between customers and developers,” says Bruin. “And because we can directly reach out to the people that downloaded Scusi, it’s easier for us to figure out what is working and what is not. We want to submit it to the Mac App Store as well, but since we built Scusi in the last two weeks, we wanted to keep things moving for now.” For iOS, there’s nothing, which is a shame since most of your messaging is probably done on your phone. It is possible to create a Shortcut that does it for you, but getting the audio file into it can be a pain. But as transcription is becoming a basic feature in more messaging apps—like WhatsApp’s feature mentioned earlier—we can hope that Apple adds it soon and perhaps even make it automatic. Then voicemail will finally be officially non-annoying.
