T he complaints started on X in July. Wispr Flow users who dictate emails, memos, and code comments multiple times a day said transcription quality had declined. The pattern continued into August. Wispr responded alongside its funding announcement with a new speech model called Canto, which the company says cuts word error rates in noisy conditions from over 30 percent to between 5 and 10 percent (Fortune, 2026).
The funding is the larger number. Menlo Ventures led a $280 million Series B that values Wispr at $2 billion, up from the $700 million valuation assigned in November (TheNextWeb, 2026). Nine months produced a tripled valuation and a product-quality complaint in the same announcement.
Flow itself converts speech into text across desktop applications rather than limiting dictation to a single program, cleaning up filler words and adjusting formatting as it transcribes (Wispr, 2026). That range is the product's whole pitch: an employee dictates into email, a CRM, a code editor, or Slack, and Flow handles the conversion the same way in each one.
The Round Priced Distribution Ahead of the Product Fix
Menlo partner Matt Kraning framed the investment around the gap between model capability and usable interfaces. Frontier labs built strong models without equally effective ways for people to talk to them (TheNextWeb, 2026). Wispr's numbers support the distribution case better than the accuracy case. Flow runs across more than 10,000 enterprises and has generated over 60 billion words of dictated text (TechCrunch, 2026). Much of that usage came from employees installing the tool themselves. Menlo noted that this spread occurred before Wispr had built a full enterprise sales organization.
That pattern matters for CIOs. A dictation layer that processes what an employee says by voice, across every application on the machine, reached thousands of companies through individual downloads, ahead of the procurement review a tool with that level of access would usually get first.
Wispr's Growth Rate Comes Without a Revenue Number
Wispr told investors revenue grew more than 150 percent in each of the last four quarters, and Menlo described the year-over-year multiple as above 30 times (TheNextWeb, 2026). Neither figure comes with an absolute dollar amount. Growth rates this high are common in early AI rounds and become harder to sustain once the early-adopter base stops being the majority of customers.
Wispr raised $25 million in November at that $700 million valuation. Ten months and $280 million later, the company is valued at nearly three times that amount.
No public figure verifies the revenue behind either number.
The Layer Wispr Wants to Own Is Also the One Shipped Free
Apple, Microsoft, and Google ship voice dictation inside their operating systems at no extra cost. Prosumer tools including Willow, Monologue, Aqua, and Superwhisper compete on price (TechCrunch, 2026). Wispr's response is to move beyond transcription toward an intelligence layer that retains context across a workday rather than treating each sentence in isolation (Wispr, 2026).
Otter and Google Recorder solve a different problem than Flow does. Otter transcribes recorded meetings and calls, with speaker labels, summaries, and action items, and has no function for dictating text live into other apps (TechBuzz, 2026). Google Recorder is a free, on-device voice memo tool built for capturing and searching audio, not for typing by voice inside a CRM or an email client. Neither product competes with Flow's core function. Wispr's Notetaker launch is the company's move onto the ground Otter already occupies.
Wispr hired Ariya Rastrow, an early Amazon Alexa team member, to lead a research group focused on that direction (TechCrunch, 2026). The company also launched a meeting note-taker called Notetaker in the weeks before the round closed. The product now expands from converting what one person chooses to dictate into a system that listens to full meetings.
That change alters what the product captures. Dictation records what an employee elects to say out loud. A meeting assistant records what everyone in the room says, not only the person who opened the app.
Unknowns and Uncertainties
Wispr has not disclosed an absolute revenue number at any point in its funding history, so the reported growth rate cannot be sized against the valuation. Canto launched as a preview alongside the funding announcement. It has not reached most of the installed base yet, so the accuracy fix users are waiting for is still ahead of them. Wispr has also not said whether Notetaker data gets different handling or training treatment than Flow's dictation data, a distinction that matters once the product captures multi-party conversations rather than one person's typed thoughts.
Garfinkle, Allie. "Wispr Talks Its Way to a $2 Billion Valuation. The Company Says Dictation's Only the Beginning." Fortune, 17 Aug. 2026, https://fortune.com/2026/08/17/wispr-2-billion-valuation-dictations-only-the-beginning/.
Mehta, Ivan. "Wispr Raises $280M at $2B Valuation as It Looks Beyond Dictation." TechCrunch, 17 Aug. 2026, https://techcrunch.com/2026/08/17/wispr-raises-280m-at-2b-valuation-as-it-looks-beyond-dictation/.
"Wispr Series B Hits $2bn as Menlo Bets the Text Box Dies." The Next Web, 17 Aug. 2026, https://thenextweb.com/news/wispr-series-b-280m-2bn-valuation-menlo-canto.
"Our Series B, and What It Means for You." Wispr, 17 Aug. 2026, https://wisprflow.ai/post/series-b.
"Wispr Flow Preps AI Meeting Notetaker to Challenge Otter, Fireflies." TechBuzz, 2026, https://www.techbuzz.ai/articles/wispr-flow-preps-ai-meeting-notetaker-to-challenge-otter-fireflies.
