TwinMind, an AI-powered app, secured $5.7 million in seed funding. Streamlined Ventures led the round, with participation from Sequoia Capital and Stephen Wolfram. This investment values the company at $60 million post-money.
The TwinMind AI app functions as a virtual second brain. It captures ambient speech with user permission, building a personal knowledge graph. This process generates AI-powered notes, to-dos, and answers from spoken thoughts and conversations.
Co-founded in March 2024 by ex-Google X scientists Daniel George, Sunny Tang, and Mahi Karim, TwinMind operates in the background. The app processes audio on-device, transcribing continuously for up to 17 hours without significant battery drain. Furthermore, it supports real-time translation across over 100 languages.
TwinMind differentiates itself from AI meeting note-takers like Otter by passively capturing audio all day. The team developed a native Swift service for iPhone, enabling extended background operation. By contrast, many competitors rely on cloud processing, which Apple restricts for prolonged background use.
Expanding AI Capabilities and Market Reach
The startup also offers a Chrome extension. This tool gathers additional context from browser activity, including email and Slack, using vision AI. Consequently, the TwinMind AI app enhances its contextual understanding beyond spoken interactions.
The daily brief
Get the free daily brief with the day's AI news, new AI startups and funding rounds every morning.
Join 64,000 other subscribers.
TwinMind introduced its Ear-3 model, a successor to Ear-2. This advanced machine learning model supports over 140 languages and recognizes different speakers. It will be available via API to developers and enterprises.
The company currently serves over 30,000 users, with 15,000 active monthly. While the U.S. represents its largest user base, TwinMind also sees traction in India, Brazil, and Europe. The TwinMind AI app targets a general audience, though professionals constitute 50-60% of its users.
TwinMind prioritizes user privacy. It does not train models on user data. Moreover, the app deletes audio recordings on the fly, storing only transcribed text locally.
