Wispr Flow Raises $280 Million and Unveils Next-Gen Voice AI

Voice AI startup Wispr Flow has reached a $2 billion valuation after closing a substantial $280 million Series B funding round. This announcement comes alongside the debut of Canto, their proprietary speech model designed to handle real-world voice interactions with greater accuracy.

“For most of the last decade, talking to computers meant asking about the weather or setting timers,” explained Wispr Flow CEO Tanay Kothari in a blog post. “Now, as voice becomes how people create critical messages and work documents, errors can disrupt their entire workflow.” The company emphasizes that accuracy is paramount—it’s not just one feature among many, but rather the foundation for whether voice AI truly transforms how we live and work.

Canto distinguishes itself with its ability to maintain high accuracy in challenging environments. According to Kothari, the model reduces error rates from over 30% to between 5-10% when handling background noise, wind, accents, or music. Across everyday usage, users can expect a 30-35% reduction in editing needed for dictations.

The Evolution of Voice AI

The funding and Canto’s launch mark a significant shift in voice AI applications—moving beyond simple commands to enabling complex work tasks through natural conversation. As PYMNTS noted recently, the industry has evolved from competing on understanding what people say to delivering tangible business value once instructions are understood.

For B2B organizations, this means compressing workflows by allowing employees to complete multiple actions within a single voice interaction—whether it’s a warehouse supervisor checking inventory while walking the floor or a field technician diagnosing equipment hands-free.