lukaszliniewicz/Pandrator

Turn PDFs and EPUBs into audiobooks; subtitles or videos into dubbed videos (including translation), and more. For free. Pandrator uses local models, including voice-cloning (instant, RVC-enhanced, XTTS fine-tuning) and LLM processing. It aspires to be a user-friendly app with a GUI, an installer and all-in-one packages.

What it solves

Pandrator is a unified workspace for creating audiobooks, subtitles, and voiceovers. It streamlines the complex process of turning text into speech or audio into text, allowing users to review, refine, and translate content across multiple AI models and providers in a single browser-based interface.

How it works

The system uses a Python backend and a Svelte frontend. It integrates a wide array of local and cloud-based AI engines for different tasks:

  • Speech Generation (TTS): Supports local engines like audio.cpp (with models such as Qwen3 TTS, Fish Audio S2 Pro, and others), Kokoro, and Silero, as well as cloud providers like OpenAI, ElevenLabs, and Google Gemini.
  • Transcription (ASR): Uses local tools like CrispASR (Whisper, Parakeet) and Qwen3 ASR, or cloud services like Azure Speech.
  • Transcription/Translation: Integrates LLMs (via OpenAI, Anthropic, Ollama, etc.) and DeepL for text cleanup, translation, and speech optimization.
  • MCP Integration: Supports the Model Context Protocol (MCP), allowing external AI agents (like Claude Code) to perform language work directly within Pandrator's workflow.

Who it’s for

  • Content Creators: People making audiobooks from documents (PDF, EPUB, etc.) or dubbing videos with synchronized speech.
  • Translators and Subtitlers: Those needing to transcribe audio/video and translate subtitles while maintaining timing and cue identities.
  • Audio Engineers: Users who want to plan recording cuts from transcripts and render edited videos.

Highlights

  • Comprehensive Format Support: Imports TXT, PDF, EPUB, DOCX, and MOBI; exports WAV, MP3, Opus, FLAC, and M4B.
  • Flexible Compute: Ability to run models locally for privacy and cost, or connect to cloud APIs for high-end performance.
  • Linguistic Tools: Includes OCR, chapter detection, glossary-aware translation, and pronunciation adjustments.
  • Agentic Workflow: Passive MCP dispatch allows existing AI assistants to process batches of work, validate results, and resume interrupted tasks.

Related

  • Project
    krillinai/OpenCreator

    An open-source AI workspace for creators that combines visual tools and AI agents to streamline the production of videos, images, voiceovers, and scripts.

  • Project
    buxuku/SmartSub

    An all-in-one open-source desktop tool for video transcription, translation, AI dubbing, and subtitle burning, featuring an AI assistant and MCP-based automation.

  • Project
    linyqh/NarratoAI

    An automated AI-powered tool for movie and TV show commentary videos that handles script writing, video editing, voiceovers, and subtitles in one workflow.

  • Project
    HAKORADev/VODER

    A professional-grade, local audio and voice processing toolkit that combines TTS, STT, voice cloning, and music generation into a single offline interface.

  • Project
    tmoroney/auto-subs

    AutoSubs is a cross‑platform desktop app that generates offline subtitles from audio/video, offering speaker labeling, many local transcription models, and direct integration with DaVinci Resolve and Adobe editing tools.