Wispr Flow is the flagship product of Wispr AI, a San Francisco company founded in 2021 that builds voice-driven text input. Flow is a cross-platform 🏷️#speech-to-text application that converts spoken language into polished, formatted text inside any application on Mac, Windows, 📝iPhone, and Android. It transcribes at roughly 220 words per minute — about four times typing speed — while editing for clarity, removing filler words, and adjusting tone to the active application. A personal dictionary learns unfamiliar terminology, a snippet library stores frequently used text, and it supports over 100 languages with automatic detection.
Wispr began as a hardware company. Founders Tanay Kothari and Sahaj Garg set out to build a non-invasive interface that would read subvocal and neuromuscular signals for silent communication, then pivoted to software as speech models matured faster than the wearable did. Flow shipped on macOS in October 2024, Windows in March 2025, and mobile shortly after. A $30 million 📝Series A led by Menlo Ventures in June 2025, with NEA and 8VC participating, was followed in August 2026 by a $280 million 📝Series B. That round funds 📝Canto, an in-house speech model the company says cuts word error rates by more than 75% in the hardest conditions (background noise, accents, mid-sentence language switching), and the 📝Wispr Advanced Interfaces Lab.
The same month, Wispr launched 📝Wispr Flow Notetaker on Mac, which records meetings with the same dictionary-aware accuracy and syncs with 📝Google and Outlook calendars; meeting notes and transcripts are reachable by AI agents over 📝MCP. Flow works from live microphone input only — it does not transcribe pre-recorded audio or video files. Its competition splits between cloud dictation services and local-first tools such as 📝Handy, which runs 📝Whisper and other models entirely on-device — the trade being Flow's cloud-side editing and formatting against on-device privacy.
Wispr Flow is one of the most-used tools in my stack, and it isn't close. I've dictated more than 1.25 million words through it, 274,014 of them in September 2026 alone, which puts me in the top 0.1% of all users. I'd never written code before 2026; now I ship software by talking to agents, and most of those words land in 📝Claude Code running in 📝iTerm2, where the custom dictionary keeps technical vocabulary intact. It records my calls too: when one ends, an agent pulls the transcript over MCP and writes the recap before I've refilled my water. If your brain moves faster than your fingers, use my referral link and get a free month of Pro.