No catalog. Your book, your voice.
Turn any book into an audiobook
Readnara reads your books aloud in Serbian or English — in a voice you choose. There is no catalog: find a PDF or an e-book, upload it and listen.
Free beta · no card needed · Serbian and English
Try another voice
The same passage in both languages, with 10 voices per language. All voices are synthetic — made by our model, not clones of real people; any resemblance is coincidental. No post-processing.
Classics library
Over 1,000 classics, ready to listen to
Alongside your own books you get a library of public-domain works: from Nušić, Sremac and Stanković to Austen, Dickens and Conan Doyle. Press play and listen.
- The school reading list in one place, for primary and secondary school
- Novels, short stories, plays and poetry, Serbian and English
- Every book shows the source it came from
Why Readnara?
Other apps sell you a catalog. We give a voice to the books you already have.
No catalog
Other apps: You only hear what a publisher chose to record.
Readnara: Find a PDF or an e-book, upload it and listen — any book.
Now, not next year
Other apps: You wait for an audio edition — which for Serbian books often never comes.
Readnara: The first page is ready a few seconds after you upload.
A voice you choose
Other apps: A narrator somebody else picked.
Readnara: Upload or record the voice you want — your own, or with permission, the voice of someone you love to listen to.
Serbian and English
Other apps: Serbian is rarely supported, and when it is, it often sounds robotic.
Readnara: Our model reads both Serbian and English books naturally.
How it works
1. Upload a book
PDF, EPUB, MOBI, FB2 or TXT — drag it in or pick it from your device.
2. Choose a voice
Pick, upload or record a voice.
3. Listen
Read along and listen; your progress is saved.
What you get
Any book, not a catalog
Upload your own PDF or EPUB and listen to it. You are not limited to someone else’s catalog like in other audiobook apps.
A voice you choose
Pick one of 22 synthetic narrators, or upload or record the voice you want to hear — your own or, with their permission, someone close to you.
Pick up where you left off
Progress is saved to the page and second. Open the book and it continues right there.
Serbian and English speech
Our model reads Serbian naturally, with correct accents — and English books just as well. The book's language is detected automatically.
Every format
PDF, EPUB, MOBI, FB2 and TXT. Scanned books go through OCR, so you can listen to them too.
Serbian and English interface
The app speaks Serbian and English; switch languages with one click.
Mini player and auto-continue
A floating mini player while you do something else, and automatic flip to the next page.
Sleep timer and phone controls
Stop reading after 15-60 minutes or at the end of a page. Pause and skip pages from the lock screen or your headphones.
Optional ambience and echo
Soft rain, sea, fireplace or gentle music under the voice, plus a small room echo. Generated in your browser, royalty-free, and never part of downloaded files.
Technology
Our own speech model
We did not plug in an off-the-shelf service. We built our own model for reading books aloud in Serbian and optimized it end to end — fewer steps and a smarter way of generating — so listening is fast and affordable for everyone.
That is why the beta can be free, and the first page is ready a few seconds after you upload.
The model is distilled from the open OmniVoice model and adapted to Serbian. The numbers are measured on our production system.
8–16
steps per sentence instead of 64 for the base model
~15×
faster than reading aloud — an hour of audio takes a few minutes
SR + EN
Serbian and English, each read by a native narrator
0
training on your books and voices — your data stays yours
For developers
Readnara API: text to speech for your app
Add natural Serbian and English speech to your site, app or service with one HTTP call. The same voices and quality as in Readnara.
- 15 minutes of generated audio free per account
- 22 synthetic voices, Serbian and English, paced for listening
- Numbers, dates and Cyrillic read correctly, no text preparation
POST https://readnara.com/api/dev/v1/tts
Authorization: Bearer rn_…
{
"text": "Dobar dan. Ovo je Readnara.",
"voice": "lib-sr-f-teodora",
"format": "mp3"
}
→ 200 audio/mpeg
X-Readnara-Remaining-Minutes: 14.92What reading looks like
The page stays exactly as in your book, the part being read is highlighted, and the player is always at hand.
- Chapters
- Bookmarks
- Read along
- Sleep timer
- M4B export
- MP3 download
Who it's for
Students
Required reading and lecture notes read aloud — study on the bus, on a walk or before an exam.
Accessibility
For people with dyslexia or low vision: listen and follow the text on the page.
People on the move
A book for the drive, the gym or the chores, without looking at a screen.
Pricing
Free beta
Every feature is free while the beta lasts.
- Up to 300 minutes of generated audio per month
- Clone your own voice
- Every book format, scanned ones included
- Reading in Serbian and English, interface in both
Paid plans later — you will be notified of any change in advance.
Frequently asked questions
Which language does it read in?
It reads books in Serbian and English; the book's language is detected automatically and you can change it. The interface is in Serbian and English too.
Is my library private?
Yes. Your books, voices and progress are visible only to your account.
Whose voice may I clone?
Only your own voice, or the voice of someone who gave you permission.
Which formats are supported?
PDF, EPUB, MOBI, FB2 and TXT. Scanned PDFs are read with OCR.
Does it work on a phone?
Yes, in the browser on your phone, tablet or computer — nothing to install.
Is it free?
Yes, during the beta, with a monthly limit of 300 minutes of generated audio.
Is the voice artificial?
Yes. All audio is generated by our AI model and is labeled as such in downloaded files.
Can I clone someone else's voice from the internet?
Not without that person's explicit permission. A voice is personal data, and recordings are protected by copyright and related rights. Report misuse at /report.
Your next book is waiting for a voice.
Create an account, upload a PDF or an e-book, and start listening in a minute.
Create an account and upload a book