NaturalReader is an AI text-to-speech platform from Canadian company Naturalsoft Ltd that reads PDFs, Word files, EPUBs and web pages aloud. It serves students and readers with dyslexia through its personal and EDU apps, and voiceover creators through a separately licensed commercial tier.
Used this tool? Rate it
Used this tool? Rate it
NaturalReader is an AI text to speech platform — an AI text-to-speech service, in the more common hyphenation — that converts written material — PDFs, Word documents, EPUB books, slide decks, scanned images and web pages — into spoken audio. The vendor's own framing on the homepage is compact and worth quoting, because it explains the unusual shape of the product line: the service is built to transform any text into natural-sounding audio, perfect for reading, accessibility and voiceover creation. Those three words — reading, accessibility, voiceover — are not marketing decoration. They correspond to three different products with three different licences and three very different price points, and confusing them is the single most common way users end up frustrated.
The product is operated by Naturalsoft Ltd, a Canadian company registered at #935-6388 No. 3 Road, Richmond, BC V6Y 0L4, Canada. Its origin story is unusually specific for a SaaS vendor: the company states it was founded in Vancouver, British Columbia, and that founder Jeff Yang developed the original software to assist with his own reading challenges during university studies. That heritage still shows in the product's design priorities — word-level highlighting, dyslexia-friendly typography, OCR for printed pages — which are the concerns of an assistive reading tool rather than of a media production suite.
NaturalReader states it has operated for roughly 20 years, serves more than 10 million users every year, and is trusted by 6000+ organizations. These are vendor self-reported figures published on the company's own About page; no independent audit of the user or organization counts was located during research, and they should be read as marketing claims rather than verified metrics.
Where NaturalReader sits in the market is best understood by contrast. Pure voice-synthesis platforms optimise for expressive output quality and API throughput for developers and studios. Screen readers optimise for operating-system-level navigation for blind users. NaturalReader occupies the middle: it is a document-first reader that happens to have a professional voice engine attached. You point it at a file or a web page, and it reads that artefact to you with the text visible and highlighted on screen. That document-centric posture explains both its strengths for students and its constraints for producers.
Note on paths: the site's marketing pages are client-rendered, and several conventional URLs (/pricing, /plans, /about, /terms) return 404 on the main host. Current pricing documentation lives on the vendor's help-centre subdomain rather than at a /pricing route.
The free tier is unusually substantial for this category. Basic voices are unlimited, and paid voice quality is sampled through small daily allowances rather than a fixed-length trial, so the free plan functions as a permanent low-tier product for people whose need is simply hearing their documents read.
Personal subscriptions, as documented in the vendor's help centre, run in three tiers: Lite at $13.90 per month or $79.00 per year, Plus at $20.90 per month or $119.00 per year, and Pro at $25.90 per month or $159 per year. Lite is restricted to Lite Voices and excludes voice cloning; Plus adds Plus and cloned voices with a shared monthly MP3 character allowance; Pro adds Pro voices and reading styles.
These figures are not consistently presented across the vendor's own surfaces. The web app page advertises Plus at $9.92/month and Pro at $13.25/month — annualised monthly equivalents — while the help centre lists $20.90 and $25.90 as month-to-month rates alongside yearly totals of $119 and $159. Both sets of numbers are current vendor material; they describe different billing bases rather than a price change, and are presented here side by side rather than reconciled. Confirm the basis at checkout.
The commercial edition is a separate product with its own pricing: Starter at $29 per user per month (500,000 credits monthly), Creator at $49 per user per month (2,000,000 credits monthly), and Team at $33 per user per month with a two-user minimum. All commercial plans include commercially licensed audio, which is the functional distinction from every personal tier.
Education licensing includes a free group licence for selected students, Lite and Plus group licences with a five-member minimum, and site licensing that the vendor lists as starting from USD $3300 for 2000 users.
Yes, with a real free tier rather than a trial. Basic Free Voices are unlimited for typing or uploading supported documents, and the higher-quality tiers can be sampled for 20 minutes (Premium) and 5 minutes (Plus) per day.
Not on a personal plan. The vendor states that audio from personal-use versions cannot be used publicly, commercially, or redistributed in any way. Publishing narration requires the separate commercial edition, whose plans include commercially licensed audio.
Quality and language reach. Plus Voices comprise 100+ natural high-quality voices across 16 languages and Premium Voices cover 8 languages, while the platform advertises 90 languages overall. Free voices are the basic engine and sound noticeably more synthetic.
PDF, TXT, DOC(X), Pages, ODT, PPT(X), ODS and non-DRM EPUB, plus JPEG and PNG images that are processed with OCR. DRM-protected e-books are excluded.
Yes, on the Plus and Pro personal tiers, each allowing up to two cloned voices. The entry-level Lite plan excludes cloning.
Personal tiers are listed in the help centre at $13.90 (Lite), $20.90 (Plus) and $25.90 (Pro) per month, with yearly options at $79, $119 and $159. Commercial plans start at $29 per user per month. Note that the web app page presents annualised equivalents of $9.92 and $13.25 for Plus and Pro, so confirm the billing basis at checkout.
EDU is a distinct product with group and site licensing, and the vendor states support for FERPA, COPPA, GDPR and New York Education Law 2-D, along with an explicit commitment that student content is never used to train AI models.
A web app, a Chrome extension, iOS and Android apps, and downloadable desktop software. The iOS app requires iOS 16.4 or later and was updated to version 9.45 in July 2026.
Because they measure different things on very different samples. The iOS app averages 4.55 from 8,831 ratings, while Trustpilot shows 1.6 from 54 reviews concentrated on billing and support experiences. Read the app rating as a signal about daily use and the Trustpilot themes as a warning about subscription administration.
Yes, indirectly. Photograph or scan the pages as JPEG or PNG and upload them for OCR; EDU also includes a dedicated OCR scanner for converting printed pages to readable text. Recognition quality depends heavily on scan clarity.
