Turn every note into a mobile-friendly, audiobook-like experience. The Obsidian Voice Plugin reads your notes aloud in natural, lifelike speech — powered by your choice of AWS Polly, ElevenLabs, Google Cloud, or Azure Speech. Listen with a dedicated player, jump between notes like chapters, change the speed on the fly, and save audio offline — with your credentials kept private in your own account.

• Highlights
• The Voice Player
• Feature Tour
• Settings
• Keyboard Shortcuts
• Choose Your Speech Provider
• Getting Started
• Connecting a Provider
• Troubleshooting & Help

• A real audiobook player — open the Voice player, see your notes as chapters, and play, skip, and repeat just like a podcast app.
• Four engines, one experience — switch between AWS Polly, ElevenLabs, Google Cloud, and Azure Speech anytime. Every feature works the same on all of them.
• Listen in seconds — turn any note into lifelike speech straight from the ribbon, a command, or the player.
• Designed for every device — the same experience on desktop, iOS, and Android, with a touch-friendly mobile player and control bar.
• Own your audio — download MP3 files, auto-embed them into your note, and keep an offline archive.
• Stay in control — adjust tempo on the fly, jump forward or back, repeat a chapter or the whole list, and watch synthesis progress in real time.
• Stay private — your credentials live in your own provider account; nothing is routed through a third party.

The Voice player is the heart of the plugin: an audiobook-style pane that turns your notes into chapters you can listen to back to back.

• Chapters from your folder — every MP3 saved next to the current note appears as a numbered chapter. Listen to a whole folder like an audiobook, and the chapter you're hearing is highlighted.
• Transport controls — previous / next chapter, rewind, play / pause, and fast-forward, with a draggable progress bar and live time display.
• Read this note — generate speech for the note you're viewing with a single click.
• Repeat modes — cycle through off → repeat one → repeat all to loop a single chapter or play through every chapter in the folder.
• Speed on the spot — nudge playback from 0.5× to 2.0× with the − / + buttons.
• Download — save the current audio as an MP3 right from the player.

On mobile, the same player opens as a full-screen pane, optimized for touch:

Tip: Open the player from the Open Voice player ribbon icon (the audio-waveform icon) or the Open the player. command.

• Start playback from the left-hand Voice ribbon icon (Voice read text) whenever you need it.

  

• Default playback reads the entire note. In Source mode, select text first and only your selection is read.

• The ribbon icon shows a refresh indicator while synthesizing and flips to a pause icon when playback is ready.

• While playback is running, press the download button to save an MP3 named after your note; the plugin embeds it right after the front matter so you can replay it anytime, offline.

  

• Prefer a hands-off workflow? Enable Auto-Save Audio to Note in settings to save and embed the MP3 automatically after every successful playback — no manual click. Off by default.

• Cached audio prevents repeat synthesis costs until your note content changes.

• Track synthesis progress with the real-time status bar indicator until playback is ready.

  

• Use rewind / fast-forward controls and on-the-fly tempo changes for quick navigation.

• Set how far rewind and fast-forward jump — configure each independently from 1 to 60 seconds in settings (defaults to 3 seconds).

• Adjust speech speed from 0.5× to 1.9× without leaving the status bar.

  

• Choose from dozens of natural voices across many languages — American, British, German, French, Spanish, Italian, Polish, Dutch, Portuguese, Brazilian Portuguese, Catalan, Swedish, Danish, Norwegian, Finnish, Japanese, Korean, Hindi, Mandarin, and more.

  

• Switch voices instantly from the status bar, or use the Switch to the next speaker. command to cycle through them hands-free.

These content toggles apply to every provider and are all off by default:

• Spell Out Acronyms — read uppercase words like NASA or API letter by letter. Off pronounces them naturally. (Applies to AWS Polly.)
• Read Code Blocks — read fenced code blocks (Mermaid, YAML, and other code) aloud. Off announces them with a short placeholder instead.
• Skip Website URLs — strip website URLs (https://… and www.…) from the spoken output while keeping the surrounding text and link labels intact. Off reads them as written.
• Auto-Save Audio to Note — automatically save and embed the MP3 after each successful playback (see Save & Play Audio Offline).

• On the Obsidian mobile app, start playback or open the player from the dedicated Voice read text and Open Voice player menu items.

  

• Control playback with the touch-friendly mobile control bar — play / pause, rewind, fast-forward, voice switching, tempo, and a progress indicator. (It stays out of the way while the full player is open.)

  

• Update credentials, validate your setup, and check voice availability directly from mobile settings.

  

• Markdown pre-processing cleans, enhances, and chunks content for reliable delivery.
• Headings, bold text, and pauses are emphasized natively on providers that support SSML (AWS Polly, Google Cloud, Azure Speech).

Everything is configured in one place: Settings → Voice. Pick a provider, enter its credentials, choose a voice, and tune playback to your taste.

Voice ships 15 commands you can bind to any hotkey. No keys are assigned by default — open Settings → Hotkeys, search for Voice, and assign whatever feels natural. (In the command palette, each command is prefixed with Voice:.)

Pick AWS Polly, ElevenLabs, Google Cloud, or Azure Speech from the Speech Provider dropdown in settings. Each provider keeps its own credentials and voice list; everything else — tempo, rewind/fast-forward intervals, downloads, auto-save, and the content toggles — works identically. After entering your credentials, press Test Credentials to confirm everything is connected.

1. Install the Voice plugin inside Obsidian (Community Plugins → Browse → Voice) and toggle it on.
2. Open Settings → Voice and pick your Speech Provider.
3. Enter that provider's credentials (see Connecting a Provider) and press Test Credentials.
4. Open any note and press the Voice ribbon icon — or open the player — to start listening.
5. Not working? See Troubleshooting & Help.

Choose one provider to start — you can switch anytime.

AWS Polly — In Settings → Voice, choose AWS Polly, select your region, paste your Access Key ID and Secret Access Key, and press Test Credentials. For a step-by-step guide to creating a dedicated AWS key, see Advanced: AWS Polly Setup.

ElevenLabs — Sign in at elevenlabs.io, open Settings → API Keys, and create a key. In Settings → Voice, choose ElevenLabs, pick a model and voice, paste the key, and press Test Credentials.

Google Cloud — In the Google Cloud console, enable the Cloud Text-to-Speech API and create an API key. In Settings → Voice, choose Google Cloud, pick a voice, paste the key, and press Test Credentials. (Don't add an HTTP-referrer restriction — see the provider notes.)

Azure Speech — In the Azure portal, create a Speech resource and copy a Key and Region. In Settings → Voice, choose Azure Speech, select the matching region, pick a voice, paste the key, and press Test Credentials.

Run into an error, see a red control bar, or need the advanced provider setup (like creating a dedicated AWS user or picking the best region)? Everything is collected in the Troubleshooting & Advanced Setup guide.