Documentary narration
A measured voice style for factual, chapter-based videos.
Local-first narration workstation
Create long-form narration without regenerating from scratch.
Caldravo splits large scripts into reviewable chunks, lets you redo only the weak sections, and exports one finished MP3 — all from your own computer.
Windows appRuns locallyNo cloud TTS creditsNVIDIA GPU recommended
Real interface
A workstation for scripts that outgrow a simple text box.
Project controls on the left, script editing in the center, and job status with chunk review on the right.
Audio examples
Hear what finished narration can sound like.
Real sample outputs for documentary, calm, and audiobook-style narration.
Sleep story narration
Slow, steady pacing for calming long-form scripts.
Audiobook chapter
Consistent narration across extended chapter sections.
The problem
Long scripts break normal TTS workflows.
Long narration projects need status, playback, retries, and a way to return later.
One weak sentence can force a full regeneration.
A wrong pause, pronunciation, or tone shift should not mean rebuilding an entire chapter.
Long jobs are hard to review and organize.
Big narration projects need chunk status, playback, notes, and a way to return later.
Cloud retries can become unpredictable.
Per-minute credit systems make experimentation stressful while the script is still changing.
Workflow
Built for long-form narration.
01
Paste your script
Drop in a chapter, documentary, course, or long narration script.
02
Generate in chunks
Split the script into manageable sections and generate them one by one.
03
Review each chunk
Listen through each section before rebuilding the final file.
04
Redo only what needs fixing
Regenerate the weak section while keeping the rest of the project.
05
Export final MP3
Combine approved chunks into one finished MP3.
Capabilities
A practical workstation, not a generic voice generator.
Long-script generation
Handle long scripts without squeezing them into a short prompt box.
Chunk review
Play, inspect, and track individual narration chunks.
Redo bad sections
Fix one weak chunk without restarting the whole script.
Project save/load
Save scripts, settings, chunks, and exports for later sessions.
Local browser interface
Use the browser as the interface while the workstation runs locally.
Built-in voice references
Use included presets or a permitted clean reference sample, depending on edition.
Final MP3 rebuild
Rebuild approved chunks into a single finished audio file.
Creator-focused workflow
Built for creators who edit, review, and return to long projects.
Use cases
Made for creators with long scripts.
Practical examples for chunk-based narration workflows.
YouTube documentaries
Turn researched scripts into sections you can review before export.
Sleep stories
Keep pacing consistent and redo only the sections that feel off.
Audiobook-style narration
Work through chapters as reviewable chunks before exporting one MP3.
Course and lesson audio
Produce clear lesson narration from structured scripts.
Podcast-style scripted narration
Review scripted episodes in parts before assembling the final narration.
Internal training material
Keep private scripts local while producing practical audio assets.
Local-first
Your scripts stay on your machine.
Keep unfinished scripts private. Generate without per-minute cloud billing. Save every chunk locally and reopen projects later when you need to continue editing.
- Windows app
- First setup may require internet and model downloads
- NVIDIA GPU recommended
- 16 GB RAM recommended
- Free Edition is best for testing compatibility
Plans
Choose the edition that fits your workflow.
Free Edition
Free
For testing Caldravo and generating basic projects.
- Local interface
- Basic long-form generation
- Chunk review
- MP3 export
- Limited saved projects
Standard Edition
$9.99 / month
For creators who want a smoother long-form workflow.
- More saved projects
- Improved project management
- More voice options
- Priority beta updates
Pro Edition — Roadmap
Coming soon
Not available today.
- Batch generation
- Subtitle/SRT export
- Audio cleanup
- Music mixing
- Workflow templates
Responsible use
Use voices responsibly.
Use only voices you own, have permission to use, or are legally allowed to use. Do not use celebrity voices, private individuals, public figures, or other creators’ voices without rights or consent.
FAQ
Questions before you download.
Will Caldravo run on my computer?
Start with the Free Edition to test setup, model compatibility, and generation speed on your hardware.
Does it need internet?
Initial setup and model downloads may need internet. Generation is designed around local project files afterward.
Where are my files stored?
Scripts, chunk audio, project settings, and exports are saved in local Caldravo project folders.
Can I redo only one section?
Yes. The workflow is chunk-based so weak sections can be regenerated without starting the whole project over.
Can I use the audio commercially?
Use depends on your voice rights, model terms, and local laws. Only use voices you have permission to use.
Start with a free test project.
Download the Free Edition, check your setup, generate a short script, and upgrade only when Caldravo fits your workflow.