Pandrator

Pandrator

Make audiobooks, subtitles, voiceovers, sounds, and music in one workspace.

Website · Tutorials with screenshots · Voice samples · Po polsku

Turn a book into something you can listen to. Transcribe a recording, translate its subtitles, or give it a new voice. Add sound effects and background music. Pandrator brings the steps together in a browser interface, with room to review and refine the result as you go.

Run speech and transcription models on your own computer, connect a cloud provider, or let your AI assistant help through MCP. Start with one workflow and add more when you need them.

The quick model guide and broader model and language matrix help you choose a voice, language, and setup that suit your computer.

Pandrator Manager installs the application and the engines you choose, then opens Pandrator in your browser. Older workspace data can still be imported. For the retired Qt desktop interface, use an older release.

Windows 10/11 and Linux desktop · x86-64 · Pandrator 0.14.2 / Manager 0.9.31

All downloads & release notes · Installation help · User guides

New in 0.14.2

• Find timing problems before export. Subtitle and recording review flag overlapping cues, overlapping word times and backwards word times without changing text or timestamps.
• Check transcription files first. A new Local model files panel checks the recognizer, speech detector, aligner and language detector separately. Administrators can reuse an existing VAD file.
• Keep captions intact. Media-edit rendering preserves whole captions with incomplete word timing, refuses cuts that would split a caption, and checks subtitle text before publishing.
• Preview the edit you will export. The 12-second mix preview follows the reviewed media-edit timeline, and exports require that exact render.
• Refresh narration length automatically. Generation resegments untouched automatic plans when the effective target changes; edited plans keep their boundaries. Breeze and VoxCPM2 targets are shorter.
• More reliable automation and long jobs. Translation branch creation is atomic and idempotent, job logs are bounded and redacted, and completed assemblies survive later progress failures.

See the 0.14.2 release notes and detailed changes.

Try your first workflow

1. Download and run Pandrator Manager. On Linux, make the AppImage executable first. Choose a parent folder; the Manager creates a Pandrator workspace inside it and opens the setup interface in your browser.
2. Install Pandrator and the engine you need. Start with one of the speech models below, or CrispASR for transcription. You can add other engines later. Docker, WSL, and a separate Python installation are not required.
3. Make something small. Choose Generate an audiobook and paste a short passage, or open Quick Transcribe to upload a clip or record your microphone. Listen, review, and save the result.

Local models download separately on first setup; speed and memory requirements vary by engine and hardware. The Manager shows the available compute options. The Windows launcher is currently unsigned, so Windows may show an unknown publisher warning; the installation guide explains this and the release checksums.

Choosing your first speech model

Choose a model that supports your text’s language, then the voice controls and hardware you need. These are useful starting points:

For AMD cards such as the RX 480, choose audio.cpp’s Vulkan backend for Qwen. Use a model package that fits your available memory and try a short passage before a long book. CPU generation is also available; speed depends on the model, package, and computer.

To reuse a designed voice, describe it → audition candidates → save the chosen sample in the Voice Library → select it for narration. With Qwen, use VoiceDesign to create the sample and Base to generate from that reference. Breeze can design a voice, save it for later generation, or clone a voice from a recording and its transcript. Record or upload your own clean reference samples when you already have the voice you want.

Breeze’s open weights and self-hosted outputs are for research and non-commercial use. For other languages or voice options, compare the broader matrix, then filter the application’s Audio models catalogue by language and capability. See providers and voices for setup and voice references.

What would you like to make?

Use the AI assistant you already have

If you use Codex, Claude Code, OpenCode, or another MCP-capable host, its model can do Pandrator’s language work directly. You do not need to configure a separate LLM provider or API key in Pandrator for this route.

In a passive MCP workflow, Pandrator prepares and tracks the work; the model in your existing conversation processes it and submits results through MCP. “Passive” describes Pandrator’s role: it does not make the LLM calls itself.

Use it to:

• Clean PDF/EPUB text before audiobook preparation, with reviewable changes to document structure and extraction artifacts.
• Correct and translate subtitles, keeping cue identities and shared terminology across batches.
• Optimize text for speech before narration, while retaining the source wording as a separate artifact.
• Coordinate a whole workflow: import a file, process the text, select a voice, generate speech, and return verified deliverables.

After connecting your MCP host, try a request such as:

Use Pandrator to correct this recording’s English subtitles and translate them into Polish. Use your own model through passive dispatch, keep a shared glossary, and export both SRT files. Preserve uncertain wording for review.

Pandrator tracks batches, validates submissions, and keeps completed work so an interrupted agent can resume. The agent must continue the tool loop; starting a passive run alone does not process it. Speech recognition and speech generation still use the engines you select in Pandrator.

Your host’s normal usage limits and costs apply. If its model runs in the cloud, the text it processes is sent there, even when Pandrator runs locally.

See the passive workflow guide, end-to-end agent guide, and the reusable workflow skill. The skill adds workflow guidance; the MCP connection supplies the tools. See skill installation to add it to your host.

Features in more detail

Audiobooks and narration

• Import TXT, PDF, EPUB, DOCX, MOBI, or pasted text. MOBI conversion needs optional Calibre.
• Prepare documents with extraction, OCR, chapter detection, text cleanup, and optional AI assistance.
• Adjust pronunciation and speech text, generate in segments, compare takes, and regenerate selected passages.
• Assign character voices and delivery directions. After changing the cast, regenerate the affected recordings and compare the saved Before and After audio versions.
• Record or upload voice references and retain their transcripts.
• Export WAV, MP3, Opus, FLAC, or M4B, with audiobook chapters, metadata, and cover art where the format supports them.
• Export a video book with timed reading text, or save timed subtitles alongside the narration.

Sounds and music

• Generate sound effects or music from a description, or import your own clips. Available controls, including lyrics, depend on the selected model.
• Save clips and alternate versions in a searchable library for reuse.
• Place effects around spoken passages and mix background music with narration or voiceovers. Adjust timing, volume, trim, and fades while retaining the original speech recordings.

See sounds and mixing and choosing a sound or music model.

Subtitles, voiceovers, and recording edits

• Transcribe audio/video with timestamps and optional speaker diarization.
• Import SRT; WebVTT, ASS, and SSA uploads are also recognized, with SRT used as the working subtitle format.
• Correct, translate, split, merge, and retime cues in a reviewable editor. Use manual edits, configured language providers, or passive MCP processing.
• Generate synchronized speech and export media with selectable audio and subtitle tracks, including burned-in subtitles.
• Plan recording cuts from a transcript, inspect and refine their boundaries, and render a reviewed edit while preserving the original media.

Quick transcription and automation

• Upload audio/video or record and preview your microphone, then copy or download TXT, SRT, or JSON without creating a permanent session.
• Transcribe through HTTP or MCP: an approved-root file path or a small base64-encoded clip can be supplied through MCP. Larger files are streamed from the host rather than placed in the model’s context.
• Quick transcripts are temporary and expire one hour after completion; save the result you want to keep.
• Use MCP for session workflows, live voice/model selection, reviewable plans, resumable work, and downloads verified by size and checksum.

Models and providers

This broader matrix covers the main speech models and providers. Language lists can differ between checkpoints, voices, and cloud models. “Listed” coverage means languages described in Pandrator’s catalogue; a model can have broader upstream claims, and pronunciation quality still varies. Open Audio models for the full list and filter by the language and voice controls you need. Speech recognition and word alignment have separate language coverage.

Local speech generation

audio.cpp is a local speech provider with several models to choose from, and the default for new workspaces. Select the model packages you want to install in the Manager, then choose an installed model in Generate audio settings. You can switch models or add more later. Available families include:

The Manager supplies model packages with their supported compute options. CPU, Vulkan, and CUDA builds are available; the Linux CUDA build is currently best-effort and has not been verified on NVIDIA hardware. Available voices and languages depend on the selected model.

These dedicated providers also remain available:

Existing workspaces keep their saved defaults and provider selections. Standalone Qwen3 TTS, Fish S2 Pro, VoxCPM2, Chatterbox, and Magpie providers remain supported for compatibility. Installed components stay manageable; uninstalled ones appear under Compatibility backends. In a saved session, Switch to audio.cpp proposes a matching model family for review. Voices and engine settings are not assumed interchangeable, and existing takes remain intact. See switching providers.

Optional RVC converts the voice of generated audio, keeping the original and converted takes available. It is a separate post-processing step. XTTS training benefits strongly from an NVIDIA GPU.

Transcription and alignment

Cloud speech and external servers

Cloud access depends on your provider account and available models. External servers are connections to separately operated services; they are not all Manager-installable engines.

Language models and translation

Use OpenAI, Anthropic, Google Gemini, OpenRouter, Groq, Mistral AI, Azure OpenAI, Google Vertex AI, or Amazon Bedrock, or connect a local LM Studio, Ollama, or OpenAI-compatible server. Model selection follows the provider’s catalogue or your configured model ID.

Correction, glossary-aware translation, document cleanup, and speech-text optimization can use a configured LLM. DeepL is also available for translation. The passive MCP route above lets your existing host model do language work instead.

The providers and voices guide covers setup, compute choices, voice references, and model imports.

Your computer, your choice of providers

Local speech and transcription engines can keep processing on your own machine. Cloud speech, language, and translation services are optional; their selected inputs are sent to those providers. Model language coverage, hardware requirements, and licences differ. Check the engine’s details before a long run.

Pandrator is designed for one owner and listens locally by default. You can also run it on a home server or GPU host; follow the remote and headless guide for setup and the privacy guide for data flows.

Go further

Help and feedback

Stuck on setup? Start with troubleshooting. Found a bug or have an idea? Open an issue with your operating system, engine/provider, and steps to reproduce it. For installation or service failures, the Manager’s Download diagnostics bundle can help; inspect it before sharing.

License

Pandrator’s source code is available under the MIT License. Third-party models, dependencies, and providers have their own licences and usage terms. Check those separately, and use source media and voice references you have the rights to use.