KINGY / DOCUMENT TO AUDIO
Your document.
Your audiobook.
Turn PDFs, Word files and EPUB books into MP3, WAV or chaptered M4B. Choose an English voice, review the text, and keep the audio.
Add your document
English OCR runs here. Review recognized text before generating speech.
Choose your narration
The first voice preview downloads the speech model, about 100 MB. Later visits can reuse the browser cache.
Add text to see a listening-time estimate.
Keep this tab open while processing. Phones may suspend work when the screen locks.
Review chapters & text
Edit names and text before creating audio. Turn off a chapter to leave it out. Uploaded DOCX headings, EPUB reading order and PDF outlines help find chapters.
Listen & keep your audio
M4B includes chapter names and navigation. MP3 and M4B encoding download an additional browser runtime, about 32 MB, on first use. Completed parts remain available when a job is paused.
Your files stay on your device.
Document reading, English OCR, speech generation and audio encoding run in this browser. Document text and generated audio are never uploaded. Kingy serves the app and processing libraries; Hugging Face serves speech-model and voice files. Those downloads disclose ordinary connection information, such as your IP address, to the file host. This tool uses no analytics, paid speech API or account.
With device saving off, text and audio are held in this tab and lost when it closes. With saving on, they are stored in this browser’s IndexedDB until you delete them; browser storage cleanup can also remove them. Recovery is local to this device and browser.
Supported formats and scanned PDFs
Text PDF, scanned PDF using English OCR, DOCX, EPUB, ODT, RTF, Markdown, TXT, HTML, CSV and JSON. Maximum input file size is 40 MB. Password-protected and DRM-protected files may need an unlocked copy. OCR can misread scans, tables and columns; always review the extracted text.
Voices and browser requirements
28 American and British English voices from the existing Kokoro 82M engine. This build provides English narration. A recent browser with WebAssembly and Web Workers is required. Processing speed and long-book capacity depend on the device. MP3 and AAC audio for M4B are encoded locally using a single-threaded runtime.
Audio formats and downloads
MP3 is a compact listening file. M4B is an AAC audiobook with embedded chapter markers. WAV preserves uncompressed speech and can be large. Apple Books and other audiobook apps handle chapter navigation; support varies by player. Keep a downloaded copy of important work.