Search...Search plugins and themes...
⌘K
Sign in
  • Get started
  • Download
  • Pricing
  • Enterprise
  • Account
  • Obsidian
  • Overview
  • Sync
  • Publish
  • Canvas
  • Mobile
  • Web Clipper
  • CLI
  • Learn
  • Help
  • Developers
  • Changelog
  • About
  • Roadmap
  • Blog
  • Resources
  • System status
  • License overview
  • Terms of service
  • Privacy policy
  • Security
  • Community
  • Plugins
  • Themes
  • Discord
  • Forum / 中文论坛
  • Merch store
  • Brand guidelines
Follow us
DiscordTwitterBlueskyThreadsMastodonYouTubeGitHub
© 2026 Obsidian

Dictation Bridge

gkipongkipon21 downloads

Dictate into the current note through a transcription gateway you control.

Add to Obsidian
  • Overview
  • Scorecard
  • Updates2

Record a short dictation in desktop Obsidian, send it to your own HTTPS gateway, and insert the transcript at the captured cursor or selection. Existing note content, title and path are not uploaded. If the note changes while waiting, the transcript stays in a recovery dialog for explicit insertion.

Community release 0.1.1. View in the Community Plugins directory. Requires Obsidian 1.14.5 or later, OS-encrypted SecretStorage, microphone access, a self-hosted gateway and a paid OpenAI API account. No hosted service or API key is supplied. Desktop only; mobile is not claimed. Provider charges and data handling are governed by the operator's OpenAI account.

Install and connect

  1. Deploy the included gateway to your own HTTPS host. Create a separate device UUID/token for each Obsidian vault/device.
  2. Install Dictation Bridge from the Community Plugins directory. For manual installation, download main.js, manifest.json and styles.css from the GitHub release; verify SHA256SUMS. Copy them into <vault>/.obsidian/plugins/dictation-bridge/, then enable the plugin.
  3. In settings, enter the complete HTTPS transcription endpoint, instance UUID and device token. This is not the OpenAI API key. No default endpoint is configured.
  4. Optionally set expected ISO language codes (empty means auto-detect), a vocabulary of up to 50 terms, and whole-line Markdown formatting commands.
  5. Open a Markdown note in editing mode and click the microphone icon or use Start / stop dictation. Assign a shortcut yourself in Obsidian Hotkeys.

Maximum recording: 4 minutes 55 seconds and 5 MiB. Stop sends the audio. Cancel discards unfinished local data; it cannot recall audio already received by the gateway/provider. Hiding the app or locking Cipherfold cancels an unfinished session. Undo uses the normal editor undo stack.

If the target note/selection changed, use Show pending transcript, choose the intended note and insert explicitly. Pending audio/text exists in memory only and is lost on cancellation, hiding, locking or exit. Retry uses the same request ID; ambiguous provider outcomes are never submitted again. See privacy.

Optional Cipherfold integration

Works without Cipherfold. If Cipherfold is present, dictation requires an unlocked compatible session and is cancelled before lock. Insertion goes through the native editor's ordinary encrypted save path. No enrollment file or private service is imported automatically.

Development

npm ci --ignore-scripts
npm run check
python3 -m unittest discover -s gateway -v

Node.js 22 or 24; Python 3.10+ and ffprobe for gateway tests. No Python package dependencies. See AGENTS.md, architecture, verification and contributing. MIT licensed; independently maintained community codebase.

HealthExcellent
ReviewPassed
About
Record short dictations in desktop Obsidian and send audio to your own HTTPS gateway for transcription. Insert transcripts directly at the cursor or selection, keep note content/title/path local, and recover pending transcripts if the target note changed. Use a self-hosted gateway with a paid OpenAI account; desktop-only and secrets stored in OS-encrypted SecretStorage.
AudioAIEditing
Details
Current version
0.1.1
Last updated
2 weeks ago
Created
2 weeks ago
Updates
2 releases
Downloads
21
Compatible with
Obsidian 1.14.5+
Platforms
Desktop only
License
MIT
Report bugRequest featureReport plugin
Author
gkipongkipon
GitHubgkipon
  1. Community
  2. Plugins
  3. Audio
  4. Dictation Bridge

Related plugins

Gemini Scribe

Allows you to interact with Gemini and use your notes as context.

WeSight

Run Claude Code, Codex, and OpenCode as local AI collaborators inside your vault.

LanguageTool Integration

Advanced grammar and spell checking, powered by LanguageTool.

Text Format

Format selected text upper/lower/capitalize.

Claudian

Embeds Claude Code/Codex and other local Agents as AI collaborators in your vault.

Smart Connections

Find related notes and excerpts while writing. Your AI link building copilot displays relevant content in graph + list view. A local embedding model powers semantic search. Zero setup. No API key.

Meta Bind

Make your notes interactive with inline input fields, metadata displays, and buttons.

Fast Note Sync

Real-time sync of your vaults across server, mobile, and web; shareable with anyone; supports REST and MCP integrations to build your personal AI knowledge base.

Thino

Quickly capture memos and display them in the sidebar with a heatmap. (Closed source)

ChatGPT MD

A seamless integration of ChatGPT, OpenRouter.ai and local LLMs via Ollama into your notes.