Notter 🔒🎙️

Notter is a local-first transcription tool built for conversations that must never leave your machine. Transcription, speaker diarization, and AI analysis all run locally using Whisper and a local LLM via Ollama. No cloud services, no uploads, no accounts. If you have a GPU, Notter uses it for faster processing. A simple web interface lets you edit speakers, summaries, and notes with full control over your data. Your audio, your insights, your hardware. Nothing else 🔒🎙️

How it works

Drop in an audio file and Notter takes it from there. The conversation is transcribed locally with Whisper, speakers are automatically separated, and a local LLM steps in to generate summaries and action items.
Everything stays on your machine, fast and private. A clean web interface lets you review, edit, rename speakers, and even chat with the transcript to dig deeper.
No cloud, no accounts, no surprises. Just your audio, turned into insight, right where it was created.

Features

Notter focuses on doing a few powerful things extremely well, all while keeping your data local and under your control.

• 🎵 Audio Transcription – Fast, GPU-accelerated transcription powered by Whisper
• 👥 Speaker Diarization – Automatic speaker separation and labeling
• 🤖 Local AI Chat – Ask questions and generate insights using a local LLM (Ollama)
• 📝 Session Management – Organize, revisit, and manage multiple recordings
• ✏️ Editable Content – Refine transcripts, summaries, action items, and speaker names
• 🔒 100% Local Processing – No cloud, no uploads, no data leakage

Build your own

If you want your own private and local transcription system, you can just build Notter in your own machine.
Check out the GitHub repo or send me a message here.