Voice technology startup Wispr has secured $280 million in Series B funding at a $2 billion valuation, aiming to replace typing as the primary way people interact with software. The round was led by Menlo Ventures with participation from Notable Capital, NEA, Neo Ventures, 8VC, and MVP Ventures, bringing the company’s total capital raised to $361 million.
The investment addresses the massive compute costs required to train proprietary voice models capable of handling real-world conditions. Wispr’s flagship product, Flow, functions as a universal voice layer that works across any text box on desktop and mobile operating systems. Unlike conventional dictation tools that produce raw transcripts requiring manual cleanup, Flow processes speech in real time — removing filler words, correcting grammar, and fixing verbal stumbles as the user speaks. The system converts rambling, informal speech into polished, professional text without post-editing.
The new funding directly accelerates development of Canto, Wispr’s proprietary speech model engineered specifically for noisy environments. Traditional speech-to-text engines perform adequately in quiet rooms but degrade sharply amid background conversations, traffic, or café noise. Canto was trained on unfiltered, chaotic audio datasets, reducing word error rates in loud settings from 30 percent to under 10 percent. This allows users to dictate lengthy emails or complex documents while commuting, working in open-plan offices, or moving through busy public spaces.
Enterprise adoption has been significant. Employees at nearly all Fortune 500 companies and over 10,000 businesses use Flow daily, with more than 60 billion words generated on the platform to date. Multilingual professionals and athletes have highlighted the system’s ability to switch between languages seamlessly without dropped words.
Wispr’s longer-term vision centers on a persistent voice layer embedded beneath every application, eliminating the need to switch between apps or type manually. As artificial intelligence advances, the company argues that physical keyboards and touchscreens may become secondary inputs, with natural speech offering faster and more accurate interaction across devices and platforms.
