What is a screen reader and how does it work?

For people who are blind or have sig­ni­fic­ant vision loss, working on a computer presents par­tic­u­lar chal­lenges. Screen readers help make on-screen content ac­cess­ible either through synthetic speech (text-to-speech) or via a tactile re­freshable Braille display.

Key Takeaways ⓘ

Screen readers make digital content ac­cess­ible through synthetic speech or Braille displays.

• They help ensure com­pli­ance with WCAG 2.2 and UK ac­cess­ib­il­ity reg­u­la­tions.
• Typical reading issues occur due to missing HTML markup like al­tern­at­ive text, th or label.
• Available solutions range from free open-source tools like NVDA to com­mer­cial programs like JAWS.

What is a screen reader?

The purpose of a screen reader is to make digital content usable re­gard­less of visual per­cep­tion. Screen readers are no longer limited to tra­di­tion­al desktop computers. They are now fully in­teg­rated into the operating systems of smart­phones and tablets, enabling ac­cess­ible use of mobile ap­plic­a­tions as well.

There is a general dis­tinc­tion between stan­dalone screen reader programs and solutions built directly into the operating system. Their scope, func­tion­al­ity, and cost vary con­sid­er­ably. In addition to free open-source solutions, there are com­mer­cial programs with pro­fes­sion­al feature sets, and licensing costs for these can reach several thousand pounds. Dif­fer­ences often stem from the use of advanced speech synthesis tech­no­lo­gies and spe­cial­ised cus­tom­isa­tion options.

Today, the quality of speech output depends less on the language itself and more on the text-to-speech engine being used. Modern neural speech synthesis supports many languages and produces much more natural pro­nun­ci­ation than earlier systems. However, dif­fer­ences remain in prosody, the emphasis of complex sentence struc­tures, and the pro­nun­ci­ation of foreign or technical terms.

The de­vel­op­ment of synthetic speech remains a central area of research in com­pu­ta­tion­al lin­guist­ics. Modern speech syn­thes­isers rely on deep learning methods and are trained on large datasets of real speech. This allows them to reproduce in­ton­a­tion, emphasis, and pauses in a realistic way. Some lim­it­a­tions persist, par­tic­u­larly in conveying emotional nuance, irony, or highly context-dependent emphasis. For screen reader users, however, today’s speech synthesis rep­res­ents a sig­ni­fic­ant im­prove­ment over the mech­an­ic­al-sounding output of earlier gen­er­a­tions.

Selection criteria for a screen reader

When choosing an ap­pro­pri­ate screen reader, com­pat­ib­il­ity with the devices used in the workplace is crucial. For example, if a Braille display is used, it must be supported by the screen reader. The same applies to the operating system as well as fre­quently used ap­plic­a­tions.

Important selection criteria include:

• Is the screen reader used primarily for reading, or also for writing?
• What types of content and ap­plic­a­tions will be used?
• What budget is available for equipping the workplace?

Depending on the specific re­quire­ments, a free solution may be suf­fi­cient, or the use of pro­fes­sion­al software may be ap­pro­pri­ate. In addition to technical aspects, factors such as ex­per­i­ence, training needs, and col­lab­or­a­tion with col­leagues also play a role. A struc­tured as­sess­ment of the workplace is therefore advisable in any case. These criteria apply not only in pro­fes­sion­al settings but also in private use.

Before pur­chas­ing paid screen reader software, it may be worth­while checking whether financial support or funding is available through Access to Work, local au­thor­it­ies, or other programs such as those provided by charities. Even though many modern screen readers are available free of charge, support or funding may still be relevant in in­di­vidu­al cases.

Ad­di­tion­al use cases for screen readers

Screen readers are no longer used only in tra­di­tion­al work settings for blind users. Although they are primarily assistive tools for people with visual impair­ments, they can also offer practical benefits for many other users. In general, screen readers are useful whenever content is better heard than read.

Typical use cases include:

• Audio playback of texts on the go: A screen reader makes it possible to have texts read aloud on a smart­phone or tablet, for example while taking a walk or using public transport. This can be more con­veni­ent than reading long content on small displays.
• Support during mul­ti­task­ing: When tasks do not require full visual attention, a screen reader can read text, in­struc­tions, or other in­form­a­tion aloud, making it easier to work ef­fi­ciently.
• Al­tern­at­ive way to consume in­form­a­tion: A screen reader can play content in a way similar to podcasts or au­diobooks, providing an ad­di­tion­al way to take in in­form­a­tion without having to read actively.
• Reducing eye strain: Reading on a screen for long periods can be tiring. Audio playback allows you to rest your eyes while still taking in the content.

These broader uses show that a screen reader is not only a tool for ad­dress­ing visual impair­ments, but also a flexible solution for ac­cess­ible, situation-based access to in­form­a­tion.

Screen readers, WCAG 2.2, and ac­cess­ib­il­ity law in the UK

Just as ramps or elevators enable people with mobility impair­ments to access buildings, ac­cess­ible websites ensure that digital content is available to as many users as possible. For people with visual impair­ments to reliably use the internet with screen readers, websites must meet specific technical and content-related re­quire­ments.

In the UK, digital ac­cess­ib­il­ity is primarily governed by the Equality Act 2010 and the Public Sector Bodies (Websites and Mobile Ap­plic­a­tions) Ac­cess­ib­il­ity Reg­u­la­tions 2018. These laws require public sector bodies to make their digital content ac­cess­ible and align with the Web Content Ac­cess­ib­il­ity Guidelines (WCAG), currently at WCAG 2.2 level AA.

The term ‘ac­cess­ible web’ does not only refer to use by people with dis­ab­il­it­ies, but also to web content that is easy to access, un­der­stand, and robust. This includes, among other things, a clear structure, plain language, and usability across different devices, operating systems, and assistive tech­no­lo­gies. Ideally, websites are designed so they can be used by people of different ages, technical abilities, and language skills.

Within the UK’s ac­cess­ib­il­ity framework, screen readers play a par­tic­u­larly important role as digital services must be im­ple­men­ted to work reliably with assistive tech­no­lo­gies. This includes, among other things, se­mantic­ally correct markup of content, a logical reading order, and full keyboard op­er­ab­il­ity of in­ter­act­ive elements.

Screen readers therefore serve not only as support for users, but also as an important testing tool for identi­fy­ing barriers in structure, nav­ig­a­tion, and content. They are crucial in ensuring com­pli­ance with the technical re­quire­ments of WCAG 2.2 and UK ac­cess­ib­il­ity reg­u­la­tions.

Typical issues when using a screen reader

When using a screen reader, issues become no­tice­able quickly if a website is not designed with ac­cess­ib­il­ity in mind. In these cases, the software may read content in­com­pletely, in the wrong order, or in a way that is difficult to follow. Even ex­per­i­enced users can then struggle to navigate the page and un­der­stand the in­form­a­tion.

Dif­fi­culties with a screen reader most often occur when reading the following elements:

• Images: If images are included without al­tern­at­ive text, a screen reader skips their content entirely. When an image is important for un­der­stand­ing the sur­round­ing text or is directly ref­er­enced, users may miss the intended meaning. Al­tern­at­ive text can be added easily in the HTML code and is read aloud by the screen reader, allowing users with visual impair­ments to un­der­stand the image. If the image also serves as a link, the al­tern­at­ive text should clearly describe the des­tin­a­tion or purpose.
• Tables: If tables are not logically struc­tured or lack clear row and column headers, reading them with a screen reader quickly becomes confusing. Many screen readers support table nav­ig­a­tion, but they cannot com­pensate for incorrect or in­com­plete semantic markup.
• Text format­ting and structure: When format­ting such as bold or italics is used to convey meaning, not all screen readers interpret it reliably. Missing or im­prop­erly coded headings further com­plic­ate nav­ig­a­tion. Clean HTML and CSS markup and a clear heading structure help users un­der­stand the content and move around a website more easily.
• Hidden content: Some websites reveal content only after a user action, such as clicking or in­ter­act­ing with an element. If this dynamic content is im­ple­men­ted poorly, a screen reader may not detect it at all or may read it in the wrong order. Complex, deeply nested struc­tures also make it harder for screen readers to interpret content correctly.
• Frames: When frames are used without clear titles, a screen reader may read isolated sections, such as nav­ig­a­tion areas, without ex­plain­ing their purpose. This makes ori­ent­a­tion difficult. Related content should be placed within a con­tinu­ous content area whenever possible, and frames are generally best avoided today.
• Forms: Form elements need clear labels so a screen reader can com­mu­nic­ate their function ac­cur­ately. All input fields and buttons should also be fully operable by keyboard to ensure ac­cess­ible use for everyone.

An overview of the most popular screen readers

Below is an overview of some of the most widely used screen reader solutions available across different operating systems. It includes both free and paid options and outlines their typical use cases.

NVDA

NonVisual Desktop Access (or ‘NVDA’ for short) is a free, open-source screen reader for Windows and is among the most widely used solutions worldwide. The program is re­l­at­ively easy to set up and is suitable for both beginners and advanced users. NVDA outputs content using synthetic speech or sends it to a connected Braille display.

NVDA is developed through donations and is updated regularly. Today, this screen reader is used primarily on Windows 10 and Windows 11 and supports a wide range of common ap­plic­a­tions.

The most important features of NVDA:

• Com­pat­ible with major web browsers, email clients, chat ap­plic­a­tions, and Microsoft Office programs such as Word and Excel
• Can be run from a single ex­ecut­able (.exe), including from external media like USB drives
• Provides extensive keyboard shortcuts for nav­ig­a­tion, control, and speech cus­tom­isa­tion
• Supports multiple speech syn­thes­isers, with eSpeak as the default and optional SAPI 4 and SAPI 5 support
• Offers output in many languages, depending on the selected speech syn­thes­iser
• Supports re­freshable Braille displays and basic mouse func­tion­al­ity

JAWS

Job Access With Speech (JAWS) is a com­mer­cial screen reader developed by Freedom Sci­entif­ic and is used primarily in pro­fes­sion­al en­vir­on­ments. For many years, the program has been regarded as the de facto standard screen reader for Windows and is available for Windows 10 and Windows 11.

JAWS is known for its stability, powerful nav­ig­a­tion features, and extensive cus­tom­isa­tion options. A time-limited trial version is available before purchase. Depending on the licensing model, costs can reach several thousand pounds. In some cases, funding or workplace ac­com­mod­a­tions through employers, edu­ca­tion­al in­sti­tu­tions, or vo­ca­tion­al re­hab­il­it­a­tion programs may help cover the expense.

The most important features of JAWS:

• Powerful speech syn­thes­isers such as Eloquence, plus support for ad­di­tion­al natural voices
• Extensive doc­u­ment­a­tion and help systems in print and Braille formats
• Support for text documents, PDFs, and, in some cases, text within images
• Fast nav­ig­a­tion using keyboard shortcuts and struc­tured lists
• High com­pat­ib­il­ity with Windows ap­plic­a­tions, browsers, and office software
• Support for numerous Braille displays

Dolphin

The Dolphin screen reader, formerly known as SuperNova, is a com­mer­cial solution for Windows systems. The screen reader is designed for blind and low-vision users who require a highly cus­tom­is­able working en­vir­on­ment. A time-limited trial version is available, after which a licence must be purchased.

Dolphin combines tra­di­tion­al screen reader func­tion­al­ity with ad­di­tion­al tools such as screen mag­ni­fic­a­tion and document analysis, making it es­pe­cially suitable for spe­cial­ised workplace needs.

The most important features of Dolphin:

• Powerful ‘Vocalizer Ex­press­ive’ speech syn­thes­iser with extensive con­fig­ur­a­tion options
• Fast keyboard nav­ig­a­tion and targeted search functions within content
• Support for PDFs and struc­tured documents
• ‘Dolphin Cursor’ for con­trolling the mouse via keyboard
• Support for numerous Braille displays
• Setup wizard for self-guided in­stall­a­tion

Narrator

Narrator is the screen reader built into Windows and is available in Windows 10 and Windows 11 without any ad­di­tion­al in­stall­a­tion. You can activate this screen reader software through the Windows settings or with a keyboard shortcut, and it provides basic functions for text output and nav­ig­a­tion.

The screen reader is primarily suitable for first ex­per­i­ences with assistive tech­no­lo­gies or for users with limited vision who only oc­ca­sion­ally rely on a read-aloud function. For permanent use in everyday work or for complex tasks, however, Narrator is only suitable to a limited extent.

The most important features of Narrator:

• Direct in­teg­ra­tion into the Windows operating system
• Simple control via keyboard commands
• Reads aloud text and basic controls
• Limited cus­tom­isa­tion options and speech quality
• No full support for complex workflows

VoiceOver

VoiceOver is Apple’s built-in screen reader for macOS, iOS, and iPadOS. It is tightly in­teg­rated into the operating system and enables blind and low-vision users to operate Apple devices using the keyboard, touch gestures, and spoken feedback.

VoiceOver is es­pe­cially effective within the Apple ecosystem and supports a wide range of languages and input methods.

The most important features of VoiceOver:

• Deep system in­teg­ra­tion without ad­di­tion­al in­stall­a­tion
• Control via keyboard, touch gestures, and trackpad
• Powerful speech synthesis, depending on operating system and voice
• Support for Braille displays
• Struc­tured nav­ig­a­tion through websites and ap­plic­a­tions

Orca

Orca is a free, open-source screen reader for Linux and part of the GNOME desktop en­vir­on­ment. The project is developed by an active community and is used by default in dis­tri­bu­tions such as Ubuntu or Fedora.

Orca is aimed at users with Linux work­sta­tions and provides basic screen reader functions combined with speech synthesis and Braille support.

The most important features of Orca:

• Tight in­teg­ra­tion with GNOME-based systems
• Uses eSpeak as the default speech syn­thes­iser
• Support for Braille displays
• Audio tutorials to help with getting started
• Setup can be complex, depending on the system con­fig­ur­a­tion

ChromeVox

ChromeVox is the screen reader built into ChromeOS and is available on Chrome­books without ad­di­tion­al in­stall­a­tion. The screen reader is designed for blind and low-vision users and can be activated through the system’s ac­cess­ib­il­ity settings or via keyboard shortcut.

Because ChromeVox is deeply in­teg­rated into ChromeOS, it is par­tic­u­larly well suited for web-based ap­plic­a­tions and cloud en­vir­on­ments. It is widely used in edu­ca­tion­al settings, where Chrome­books are commonly deployed, and allows users to navigate websites, documents, and system in­ter­faces using keyboard commands and spoken feedback.

The most important features of ChromeVox:

• Built directly into ChromeOS and ready to use without in­stall­a­tion
• Extensive keyboard nav­ig­a­tion with struc­tured commands for headings, links, and form elements
• Optimised for web ap­plic­a­tions and browser-based content
• Spoken feedback and optional Braille display support
• Quick ac­tiv­a­tion and de­ac­tiv­a­tion via keyboard shortcut
• In­teg­ra­tion with Google Workspace ap­plic­a­tions such as Docs, Sheets, and Gmail

Speechify and Chirpy

Speechify for Google Chrome and Chirpy for Mozilla Firefox are simple browser ex­ten­sions for reading web content aloud. They are aimed at users who oc­ca­sion­ally want to listen to text, but they are not a re­place­ment for full-featured screen reader software.

The most important features of the browser ex­ten­sions:

• Quick in­stall­a­tion and easy to use
• Suitable for simple web content
• Limited speech quality and func­tion­al­ity
• Not suitable for users with severe visual impair­ments

TalkBack

TalkBack is the default screen reader for Android and is activated via the system settings under ‘Ac­cess­ib­il­ity’. The screen reader app is free and deeply in­teg­rated into the operating system.

TalkBack uses speech output, audio signals, and vibration feedback to enable ac­cess­ible use of smart­phones and tablets.

The most important features of TalkBack:

• Standard solution for Android devices
• Good system in­teg­ra­tion and ongoing de­vel­op­ment
• Operation via gestures and key com­bin­a­tions
• Per­form­ance depends on hardware and device model

TalkBack runs con­tinu­ously in the back­ground, which can cause no­tice­able slowdowns on some devices. For that reason, it works best on a higher-per­form­ance smart­phone. Delays can be es­pe­cially chal­len­ging for blind users, as it is not always im­me­di­ately clear whether the device is still pro­cessing or if an input error occurred. Although TalkBack is a powerful screen reader app, it depends on adequate device per­form­ance to operate smoothly.

Screen reader overview