Package Exports

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (samuraizer) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

Samuraizer

Local-first CLI for meeting transcription and analysis.

Samuraizer processes audio recordings and generates:

transcript
summary
action items
decisions
report

All processing is done locally using tools like Whisper and Ollama

✨ Features

🎙 Transcribe audio recordings (Whisper)
🧠 Generate summaries (local LLM via Ollama)
✅ Extract action items
📌 Extract decisions
⚡ Resume processing (skip already processed steps)
🔧 Simple CLI + config system
🔒 Local-first (no cloud required)

📦 Installation

npm install -g samuraizer

⚙️ Prerequisites

Make sure you have installed:

Node.js >= 20
ffmpeg
whisper-cli (whisper.cpp)
Ollama
Start Ollama
ollama serve
ollama pull qwen2.5:14b

🚀 Quick Start

samuraizer init
samuraizer process meeting.m4a

⚙️ Configuration

Samuraizer uses a global JSON config file.

Initialize config

samuraizer init

Config location

macOS: ~/Library/Application Support/samuraizer/config.json
Linux: ~/.config/samuraizer/config.json
Windows: %AppData%/samuraizer/config.json

View config

samuraizer config get

Example config

{
  "model": "qwen2.5:14b",
  "ollamaBaseUrl": "http://127.0.0.1:11434",
  "whisperCommand": "whisper-cli",
  "ffmpegCommand": "ffmpeg",
  "ffprobeCommand": "ffprobe"
}

🧩 Config fields

model — LLM model used for analysis (summary, action items, decisions)
ollamaBaseUrl — URL where Ollama is running
whisperCommand — Command used to run Whisper
ffmpegCommand — Command used for audio processing
ffprobeCommand — Command used for audio inspection

🧪 Commands

Show help

samuraizer --help

Show version

samuraizer --version

🚀 Full pipeline

Process an audio recording:

samuraizer process meeting.m4a

Show detailed metadata after processing:

samuraizer process meeting.m4a --verbose

Recompute all steps even if outputs already exist:

samuraizer process meeting.m4a --force

You can combine flags:

samuraizer process meeting.m4a --verbose --force

🎛 Individual commands

Normalize audio to Whisper-compatible WAV:

samuraizer normalize input.m4a output.wav

Summarize a transcript file:

samuraizer summarize transcript.txt

Extract action items from a transcript file:

samuraizer actions transcript.txt

Extract decisions from a transcript file:

samuraizer decisions transcript.txt

⚙️ Configuration commands

Create the default config file:

samuraizer init

Print config file path:

samuraizer config path

Print resolved config as JSON:

samuraizer config get

📂 Output

output/<file-name>/
  transcript.txt
  summary.txt
  action-items.json
  decisions.json
  report.txt

🔁 Resume Behavior

Samuraizer skips already processed steps.

Use --force to rebuild everything.

⚠️ Common Issues

Ollama not running

ollama serve

ffmpeg not found

macOS:

brew install ffmpeg

Linux:

apt install ffmpeg

📄 License

ISC