Wispr’s $280 Million Bet Is That Voice Can Move Beyond Dictation

Wispr is trying to turn voice from a faster way to type into a wider workplace interface, spanning dictation, meeting notes and hardware-enabled input. Its new $280 million Series B gives that effort a $2 billion valuation—and makes speech accuracy the central product test.

Menlo Ventures led the round, which brings Wispr’s total funding to $361 million. Existing backers including Notable Capital, NEA, Neo Ventures and 8VC participated, while Acrew, Forerunner, Goodwater, Peak XV and Together Fund joined as new investors.

The valuation has climbed quickly. Wispr raised $25 million at a $700 million valuation less than a year ago, according to Entrackr. Its latest financing therefore gives the company considerably more capital to pursue product expansion and speech-AI development at the same time.

The capital arrives as Wispr broadens the scope of its flagship product, Wispr Flow. The app turns spoken words into formatted text, removes filler words and adapts to a user’s vocabulary; Wispr says the platform is used by more than 10,000 businesses.

From dictation to meeting notes

Wispr has released a meeting notetaker that can show summaries and action items, placing it alongside services such as Granola, Fireflies and Read AI. The product has room to add integrations that would let it update other tools or create documents and email drafts, but those functions were not described as current capabilities.

Canto is the immediate reliability test

Alongside the financing, Wispr previewed Canto, its first proprietary speech model. The company says Canto is built for difficult recording conditions, including background noise, wind, heavy accents and music—settings where it says word-error rates can fall from more than 30% to between 5% and 10%.

Wispr also expects Canto to reduce the share of everyday dictations that need editing by 30% to 35%. That is a company projection rather than an independent benchmark, and it follows user complaints in recent weeks about a decline in the quality of Wispr Flow’s dictation output.

The model is central to the company’s stated spending priorities. Tanay Kothari said accuracy will receive the first and largest share of the new round because errors matter more when people use voice to write work messages and documents. Canto’s results are therefore more than a feature claim: they are the operational premise for Wispr’s wider product push.

Where Wispr is extending its reach

• Meetings: its notetaker currently provides summaries and action items.
• Devices: Wispr is partnering with hardware makers including the Oasis ring so users can dictate without speaking loudly.
• New interfaces: Wispr launched Interface Labs under Chief Scientist Ariya Rastrow to explore human-computer interaction interfaces.

The expansion has a demanding market test

Wispr has released its dictation app on Android and expanded go-to-market teams in India and the U.K. India is one of its top three organic markets, according to Together Fund, which said roughly 60% of Wispr’s dictation is in non-English languages.

Wispr’s founders are Tanay Kothari and Sahaj Garg. The company also put Ariya Rastrow, described as an early member of the team behind Alexa, in charge of Interface Labs last month. That lab is intended to explore new human-computer interaction interfaces, extending the company’s interest beyond dictation software alone.

The challenge is no longer simply transcribing speech. Wispr is entering a crowded dictation market while also taking on existing meeting-note products. Its valuation now rests on a broader proposition: that people will trust voice input for work that needs clean text, useful follow-up and less editing than today.