Audoc: Document to Audio

Turn documents into structured listening sessions that explain more than plain text. Audoc imports PDF, Word, PowerPoint, Excel, OpenDocument, EPUB, HTML, RTF, Markdown, text, CSV, and images. It preserves headings, text, tables, charts, diagrams, and illustration locations where the source format exposes them, so supported visuals are heard where they actually appear. • Narrated and Plain Read modes • Table narration with headers and rows • Chart trends, comparisons, high points, and low points • Per-visual retry and complete-script regeneration • Automatic document-language detection • Matching voice selection for original and translated text • Screen-off, foldable, background, lock-screen, and notification playback controls • Saved translation progress with pause, resume, retry, and restart-all Audoc is local by default. Parsing, saved scripts, local AI, and offline neural voices remain on your device. Optional Gemma 4 E2B and E4B downloads provide private translation and richer visual understanding. Optional Supertonic, Kokoro, and Chaowen packs add compatible offline neural voices. For non-confidential documents, you may enter your own Gemini or OpenAI API key. Cloud processing is disabled for each document until you explicitly allow it. Your provider’s charges, limits, data practices, and terms apply. Imports are limited to 256 MB, with tighter safety limits for individual text, image, and archive components. Optional models are large and performance depends on device RAM, storage, thermal conditions, language, and document complexity. Legacy DOC, PPT, and XLS files may expose less structure than DOCX, PPTX, and XLSX.