DevWhisper
A native Windows local AI meeting assistant. It captures system audio and the microphone through WASAPI loopback, transcribes with whisper.cpp and extracts structured tasks using a local Ollama code model. Data never leaves the local machine, targeting freelance developers in NDA-sensitive settings.
Timeline (last 90 days)
DevWhisper is a native Windows desktop application that simultaneously captures system speaker audio and microphone audio via WASAPI loopback to record two-sided conversations.
DevWhisper uses whisper.cpp (via the pywhispercpp bindings) for local audio transcription.
DevWhisper uses the qwen2.5-coder:7b model on Ollama to extract structured technical tasks, priority tags, and assignees, and to draft client follow-up emails.
DevWhisper uses Pydantic together with Ollama's JSON schema constraints to ensure reliable output structure.
DevWhisper stores transcripts in a local SQLite database and supports full-text search via FTS5.
DevWhisper deletes temporary audio files from disk immediately after processing completes, and all data flows only through localhost and system memory.
A senior developer and tech community leader used DevWhisper to process recordings of technical project sync meetings and reported that the capture accuracy for technical action items (such as pin configurations and API endpoints) was noticeably higher than with general-purpose note-taking tools.
All Facts (7)
DevWhisper is a native Windows desktop application that simultaneously captures system speaker audio and microphone audio via WASAPI loopback to record two-sided conversations.
50%UnverifiedDevWhisper uses whisper.cpp (via the pywhispercpp bindings) for local audio transcription.
50%UnverifiedDevWhisper uses the qwen2.5-coder:7b model on Ollama to extract structured technical tasks, priority tags, and assignees, and to draft client follow-up emails.
50%UnverifiedDevWhisper uses Pydantic together with Ollama's JSON schema constraints to ensure reliable output structure.
50%UnverifiedDevWhisper stores transcripts in a local SQLite database and supports full-text search via FTS5.
50%UnverifiedDevWhisper deletes temporary audio files from disk immediately after processing completes, and all data flows only through localhost and system memory.
50%UnverifiedA senior developer and tech community leader used DevWhisper to process recordings of technical project sync meetings and reported that the capture accuracy for technical action items (such as pin configurations and API endpoints) was noticeably higher than with general-purpose note-taking tools.
50%