aside

Transcribe meeting audio and generate structured Obsidian vault notes.

33|3|Updated Feb 28, 2026
One-click install
npx skills add https://github.com/jshph/aside --skill aside-jshph
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aside
Source: https://github.com/jshph/aside/tree/main/skills/aside
Command: npx skills add https://github.com/jshph/aside --skill aside-jshph

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires fluidaudiocli, diarize, numpy, scikit-learn, ffmpeg, ffprobe, and includes scripts (resource) components.

What problem does it solve?

It turns meeting audio and messy real-time notes into a clean, structured Obsidian vault note without forcing you to leave your existing workflow.

Core Features & Use Cases

  • Local transcription: Converts audio to text on your machine, including stereo aside recordings and mono voice memos.
  • Timeline alignment: Matches timestamped memo lines with transcript segments so the conversation can be reviewed on a shared timeline.
  • Vault-connected distillation: Uses Enzyme-backed vault search to weave existing notes, tags, and people into the final note.
  • Flexible session handling: Supports align-only runs, diarization for multi-speaker audio, and publishing the result back into the vault.

Quick Start

Use the aside skill on the named session to transcribe the audio, align it with the memo, and draft the vault note.

Frequently Asked Questions about aside

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I transcribe meeting audio and organize it into Obsidian vault notes?

aside transcribes meeting audio locally and organizes the results into vault-connected notes by aligning transcript segments with timestamped memos and weaving in existing vault tags and people.

Can I use diarization to separate speakers when transcribing multi-speaker recordings?

Yes, aside supports optional diarization for multi-speaker audio, enabling speaker-aware cleanup so that group discussions, interviews, and calls are accurately attributed in the final vault note.

Does aside require local speech transcription tools to process voice memos and lectures?

Yes, aside requires local speech transcription using dependencies like fluidaudiocli and diarize, along with ffmpeg and ffprobe, to convert mono voice memos and stereo recordings into text on your machine.

How do I align timestamped real-time notes with transcript segments on a shared timeline?

aside performs timeline alignment by matching timestamped memo lines with transcript segments, allowing you to review the conversation and your notes together on a unified timeline within the vault note.

What is the best way to connect existing notes and tags to newly transcribed meeting audio?

aside uses Enzyme-backed vault search to distill the transcript, automatically weaving relevant existing notes, tags, and people into the final structured note so your meeting output stays connected to your vault.

Can I run align-only processing without generating a full vault note?

Yes, aside supports flexible session handling including align-only runs, which match memos to transcripts without proceeding to the full structured note generation and vault publishing step.