Features
Assistivox is under active development. This list reflects what works today.
- Voice dictation, with spoken commands for punctuation, capitalization, and formatting
- Voice editing commands: select, delete, capitalize, undo, move by word or sentence, and more
- Text-to-speech playback of the open document, sentence by sentence, with the current sentence highlighted as it speaks
- Keystroke echo: the editor speaks each key as you type and reads back the last word when you finish it
- A persistent on-screen keyboard tray for document navigation and spell checking, usable without a physical keyboard
- Spell check with a personal added-words dictionary you can review and remove words from
- Opening documents:
- PDF, with text extracted directly from born-digital files, or sent to OCR automatically if the file has no readable text
- Word (.docx)
- PowerPoint (.pptx)
- Plain text
- Exporting documents to Word (.docx), preserving tables, lists, and headings
- Optical character recognition (OCR), available two ways:
- Local, on-device OCR that works without an internet connection
- AI-based OCR (Gemini Flash 2.5 through OpenRouter) for harder documents, using your own API key
- Camera Scan: point your device camera at a printed page and extract the text
- Clipboard Reader: paste text in and hear it read aloud immediately
- Scratch Pad: a lightweight voice-driven notepad
- Audio Player: play audio files with bookmarks, per-bookmark notes, and quick controls for seeking and jumping to marks
- Lecture Studio: record audio or screen lectures, review and trim them, split and reorder segments, and export the result as a project file
- Dropbox integration: open and save files directly to Dropbox
- Encrypted settings backup: export your API keys and settings to a passphrase-protected file, and restore them on any device
- Installable as an app (PWA) on Android, iPhone, and desktop, with no app store required
- Bring your own AI: connect your own OpenRouter or Cloudflare account for AI-assisted document conversion and OCR
Assistivox runs in your browser. Your documents and API keys stay on your device.