Some of the efficient use instances to have emerged from fashionable AI fashions is their capability to know the human voice higher, which has led to extra correct dictation instruments. One such Mac device, Willow, is launching an iOS app to let customers sort with their voice throughout all apps.
Much like Wispr Stream, Willow’s keyboard transcribes your voice and likewise codecs the message in line with the context. The app, which helps over 100 languages, permits you to outline your individual vocabulary phrases or create completely different writing kinds throughout app classes like work, messaging, e-mail, and others.
One key benefit of Willow’s app is that there’s a full keyboard obtainable to sort, whereas Wispr Stream solely affords a numeric keyboard. This helps with fast edits to switch some phrases or sentences quite than typing them out. Plus, you don’t have to change to a different keyboard if you need to sort some phrases as a substitute of talking them.
Willow was based by Allan Guo and Lawrence Liu, who attended Stanford and dropped out to construct a startup. The duo acquired into Y Combinator in the summertime of 2024 with a unique set of co-founders (who’ve now left the corporate), and their preliminary concept was in healthcare — particularly to construct software program to handle assisted residing services. Nonetheless, the thought didn’t stick, and for a 12 months, the corporate tried to construct various things.
“Whereas we had been engaged on healthcare options, we observed that docs have voice AI scribes that report conversations with sufferers and create paperwork for follow-ups. I had a variety of conversations with docs; they stated these notetakers had been very helpful to them. This nudged us to construct a voice AI device for information staff and different customers,” Guo advised TechCrunch over a name.
Guo added that he selected to not construct one other AI notetaker as a result of he felt that the market was saturated. As an alternative, the dictation area was extra interesting as a variety of communication takes place exterior conferences. He stated that the corporate makes use of a collection of fashions and focuses closely on tuning its text-to-text pipeline, which is predicated on Meta’s Llama fashions, for formatting and personalization
The startup then moved its YC batch to Spring 2025 to launch the product. The corporate stated that because the launch, it has grown 50% month over month by way of customers. It additionally has enterprise prospects like Uber, Heidi Well being, and Zego utilizing the device for dictation with options like customized staff vocabulary.
Techcrunch occasion
San Francisco
|
October 13-15, 2026
Willow has raised $4.5 million from Field Group, Y Combinator, Burst Capital, and angel buyers equivalent to HubSpot’s Dharmesh Shah; Gusto’s Tomer London; Reddit co-founder Alexis Ohanian; ex-Yelp COO Kipp Bodnar; Opendoor CEO Kaz Nejatian; Adam Guild, co-founder and CEO of the restaurant software program firm Proprietor; and Instacart co-founder Max Mullen.
Mullen stated that he believes that essentially the most wonderful person interfaces sooner or later are going to be voice-led.
“I used to be actually impressed with Allan’s imaginative and prescient round not simply creating an awesome dictation app however, long run, aiming for creating an interface that may management your pc,” he stated. “When I’m writing by Willow, I must make fewer edits than I needed to with a pc’s built-in dictation system.”
Mullen identified that one among his favourite workflows was that on desktop, it’s also possible to use the “Hey Willow” assistant to instruct the app to put in writing issues like e-mail replies in your individual voice.
The startup competes with different gamers like Wispr Stream, which has raised greater than $56 million in funding to this point; Monologue, which is a part of Each’s subscription bundle; and the YC-backed startups Aqua, Talktastic, Superwhisper, and Betterdication.
Guo stated that within the coming months, it plans to do platform growth with Home windows and Android, together with higher personalization to cut back the variety of handbook edits customers must do put up dictation.
