FlowFlow, voice notes with on-device RAG in Rust for iOS (github.com)

FlowFlow, a new mobile app for iOS, has been launched to streamline voice note-taking with advanced AI capabilities, all built on Rust and Dioxus 0.7. This local-first app combines text and voice note editing in a unified interface, offering features like real-time audio visualization during dictation, automatic note transcription via the Soniox REST API, and AI-generated tagging through integration with OpenAI and Anthropic models. Notably, the app supports a Retrieval-Augmented Generation (RAG) chat function, enabling users to query their notes and receive AI-driven responses with embedded source references. This release is significant for the AI/ML community as it showcases the potential of RAG techniques in enhancing personal productivity through intelligent note management. The app operates on a local infrastructure using SQLite and LanceDB for semantic search, ensuring data privacy while leveraging powerful AI models. With features such as real-time search, folder management, and markdown-rendered responses, FlowFlow demonstrates a robust integration of modern AI functionalities into everyday tasks, potentially setting a new standard for mobile applications focused on information management and productivity.