run-llama

run-llama / audio-kb

Public

Record, store and search audio notes from the terminal, powered by LlamaParse and Gemini Embeddings 2

16
0
100% credibility
Found Mar 11, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A terminal tool for recording or uploading audio notes, converting speech to text, storing it locally, and searching semantically by natural language queries.

How It Works

1
🔍 Find Your Voice Note Buddy

You discover a simple computer tool that lets you save spoken thoughts and find them later by asking questions.

2
📥 Add It to Your Computer

You grab the tool onto your machine with a quick download command, and it's ready to use.

3
📓 Start Your Private Notebook

You launch a safe spot on your computer to keep all your voice memories organized and secure.

4
🔗 Link Smart Understanding

You set simple preferences so the tool can turn your speech into words and spot similar ideas.

5
Capture a Voice Note
🎙️
Speak Live

Hit enter, talk your ideas, hit enter again to wrap up.

📁
Pick a File

Choose an audio file from your files to add.

6
✨ Watch It Transform

Your words become searchable notes, neatly filed away in your notebook for easy recall.

7
🔎 Ask and Discover

Type a question like 'What did I plan for dinner?' and see matching notes pop up.

🎉 Memories at Your Fingertips

You effortlessly pull up exactly what you said before, feeling organized and in control.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 16 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is audio-kb?

audio-kb is a Python CLI tool for recording, transcribing, and semantically searching audio notes right from your terminal. Drop in an MP3 or hit record for live audio—like an audio record GitHub project with built-in storage—then it transcribes via LlamaParse, embeds chunks with Gemini Embeddings 2, and indexes them in a local SurrealDB vector store. Search queries pull relevant snippets by cosine similarity, turning rambling voice memos into a searchable knowledge base without leaving the command line.

Why is it gaining traction?

It stands out by combining dead-simple terminal recording (press enter to start/stop) with pro-grade AI parsing and local vector search, sidestepping clunky apps or cloud dependencies. Devs dig the `audio-kb process` for instant upload/transcribe/embed and `audio-kb search` for natural-language hits, complete with JSON output or Markdown previews. No fuss with audio KB/s tweaks or bitrate changers—it's plug-and-play for quick audio KB reducer workflows.

Who should use this?

Terminal diehards capturing ideas during code sprints, podcasters building a personal audio KB, or researchers needing searchable call records without apps. Ideal for backend devs or data scientists ditching notebooks for voice-driven note-taking, especially if you're already in LlamaIndex or Gemini ecosystems.

Verdict

Worth a spin for early adopters—solid README, easy uv install, and functional CLI—but with just 10 stars and 1.0% credibility score, treat it as alpha: test on non-critical notes and watch for bugs. Great prototype for audio KB on GitHub, but production users should wait for more polish.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.