What problem does it solve?
Manually transcribing voice messages from Telegram or local audio files and organizing them into your Obsidian vault is tedious and time-consuming, creating friction for capturing quick thoughts and ideas.
Core Features & Use Cases
- Multi-format audio transcription: Supports OGG, Opus, m4a, mp3, wav, and flac audio files, with options for private local Whisper.cpp transcription or fast OpenAI Whisper cloud transcription.
- Auto-organized Obsidian notes: Automatically generates timestamped, slugged Inbox notes that include the full transcript, audio duration, and source file path, eliminating manual filing work.
- Use case: When you receive a Telegram voice note with a spontaneous project idea, this skill transcribes it and saves it directly to your Obsidian Inbox so you can reference and act on it later without any manual steps.
Quick Start
Use the ingest-voice-note skill to transcribe the attached Telegram voice message and save the transcript to your Obsidian Inbox.