Wispr closed a $280 million Series B led by Menlo Ventures, valuing the voice-input startup at $2 billion, the company said Monday. The round comes less than ten months after its previous raise, a $25 million check from Notable Capital that TechCrunch reported on last November. Wispr’s total funding now stands at $361 million, and the new money is meant to fund a shift the company badly needs: from a dictation app into something closer to a general voice interface.
That shift is the real story here. Dictation, converting speech to text, is a feature Apple, Google, and Microsoft already build into their operating systems for free, and OpenAI and other model makers bundle voice modes into their chat products at no extra charge. A startup priced at $2 billion cannot survive on a feature its users’ phones already ship. Wispr’s answer is to expand into meeting notes, hardware partnerships, and what it calls a broader interface layer for talking to computers.
On the funding mechanics: existing backers Notable Capital, NEA, Neo Ventures, 8VC, and MVP Ventures all put in more money this round. New names joining include Acrew, Forerunner, Goodwater, Peak XV, Together Fund, and PLUS Capital. Neither TechCrunch’s report nor Wispr’s announcement discloses whether the $2 billion figure is pre-money or post-money, nor does either disclose revenue, active users, or retention. The error-rate figure below is the only performance number attached to this raise, and its absence elsewhere is worth noting on its own: a $2 billion price with no disclosed usage or revenue data is itself part of the story.
That one figure concerns quality, not growth. Several users had complained in recent weeks about a drop in output quality from Wispr Flow, the core dictation product. Wispr says a new model called Canto, launched alongside the funding, cuts its error rate from 30% to under 10%. That is a defensive fix for an existing product, not evidence the expansion plan is working.
The expansion itself has several pieces. Wispr’s meeting notetaker, which surfaces summaries and action items, now competes directly with Granola, Fireflies, and Read AI, though it does not yet integrate with other tools to draft documents or emails the way some rivals do. The company has shipped an Android app since November and grown its go-to-market teams in India and the United Kingdom. It is also partnering with hardware makers, including the Oasis ring, so users can dictate without speaking aloud.
The most telling piece is organizational. Last month Wispr launched Wispr Interface Labs, run by Ariya Rastrow, who worked on Amazon’s Alexa in its early years. Alexa reached mass adoption in consumer hardware but never became the default way people operate a computer, a caution built into the résumé of the person Wispr just hired to try again. Hiring for that ambition signals Wispr sees dictation as a bridge, not a destination, though TechCrunch’s report gives no timeline or product detail for what the lab will ship.
Dictation apps such as Willow, Monolouge, Aqua, and Superwhisper are already undercutting Wispr on price, and free entrants keep arriving for prosumers. For operators evaluating voice tools, the practical move over the next quarter is to treat Wispr’s transcription accuracy as table stakes and watch Wispr Interface Labs for whether it ships a product that free, OS-level voice features genuinely cannot replicate.
TechCrunch, reported by Ivan Mehta, published August 17, 2026.
