Char provides a local‑first AI notepad that captures system audio during meetings, transcribes it on‑device or via selected STT services, and generates AI‑driven summaries and contextual Q&A. All output is stored as plain markdown files with end‑to‑end encryption, giving users secure offline access and easy integration with version control or knowledge bases. The platform supports plug‑and‑play LLM and STT providers, customizable meeting templates, and extensible APIs for automation.
Funding
Funding not disclosed
Founders
Product
Problem
Professionals and teams often rely on manual note‑taking or third‑party transcription services that store audio and text in proprietary clouds, creating security risks, vendor lock‑in, and fragmented workflows. This makes it difficult to capture complete, searchable meeting records while maintaining data privacy and compliance.
Solution
Char delivers a local‑first AI notepad that captures system audio in real time, generates on‑device speech‑to‑text, and augments the raw transcript with AI‑driven summaries and contextual Q&A. Users retain full ownership of their data as plain markdown files, which can be synced or kept offline at their discretion. The platform supports plug‑and‑play integration with any Speech‑to‑Text (STT) or Large Language Model (LLM) provider—cloud or self‑hosted—allowing organizations to align the stack with their security policies. Customizable meeting templates and a floating control panel streamline workflow across sales, product, engineering, and recruiting use cases, while keyboard shortcuts and API hooks enable deep automation.
Target Audience
The primary users are knowledge workers—sales, product, engineering, and recruiting professionals—who conduct frequent virtual meetings and require secure, searchable, AI‑enhanced notes. It is also suited for regulated industries (e.g., healthcare, finance) that mandate on‑premise data handling.
Features
- System‑level audio capture that works with any meeting application, eliminating the need for meeting bots or external recording services.
- Real‑time transcription pipeline using on‑device Whisper models (Apple Silicon) or optional cloud STT providers, with end‑to‑end encryption of audio streams.
- AI summarization engine that merges user notes and transcript, powered by user‑selected LLMs (OpenAI, Anthropic, Gemini, Mistral, or local Ollama models).
- Interactive AI chat interface that queries the live transcript via natural language, returning context‑aware excerpts without leaving the app.
- Plain‑text markdown export and versioned storage, enabling seamless integration with version control, knowledge bases, or downstream analytics pipelines.
- Extensible plugin architecture (hooks, REST API, and developer SDK) for custom actions such as auto‑tagging, CRM sync, or compliance audit logging.
- Fully open‑source codebase (MIT license) with transparent security audits and community‑driven extensions.
- Cross‑platform template library for sales calls, product road‑mapping sessions, sprint retrospectives, and interview debriefs, configurable via JSON schema.