Did you know that a student can listen to every word of a digital textbook and still be using the wrong accessibility tool? A button marked “Read aloud” may help with a paragraph, but it may not tell the student where a heading begins, which field requires an answer or whether a link leads away from the worksheet. Another button marked “Dictate” does almost the opposite: it listens to the student and makes written text.
Vocabulary matters because text-to-speech, screen reader and dictation describe three different learning jobs. Knowing which job is needed helps a child, parent or teacher choose a suitable tool, give accurate instructions, check the result and ask for accessibility support without embarrassment. The aim is not to replace reading or writing practice; it is to remove an avoidable barrier so meaningful learning can take place.
Three tools that sound alike but do different work
- Text-to-speech (TTS): turns digital text into audible speech. The student is primarily listening to existing text. A simple read-aloud feature may speak selected paragraphs, and some tools highlight words as they are spoken.
- Screen reader: a more comprehensive assistive technology that can announce controls, headings, links, fields and other interface information as well as reading text. It helps people navigate and operate accessible apps and websites, often through keyboard, touch or other commands.
- Speech-to-text or dictation: turns spoken words into written text. The student is primarily composing or entering text by speaking; the output needs checking.
- Speech control or voice access: may let the learner operate buttons, menus and other controls through speech. It overlaps with dictation in some products but is not the same as merely generating a transcript.
The Web Accessibility Initiative at W3C explains the different jobs of text-to-speech and screen readers and speech recognition. Notice its important distinction: recognising spoken words is different from identifying a particular speaker. Those two meanings can otherwise be confused when an app advertises “voice recognition”.
A short classroom story: three students, one Science passage
The following is a fictional teaching example, not a report of real student outcomes. A Primary 5 class is reading: “As water evaporates from a wet surface, the liquid changes into water vapour. Evaporation may occur below the boiling point.” The unfamiliar word is evaporates. The task is to explain the idea in ordinary language and give an example.
Maya wants to hear unfamiliar scientific words
Maya can see and operate the page, but listening while following a highlighted line helps her process a dense paragraph. She selects text-to-speech, listens once, pauses after evaporates, and rereads the printed sentence. Next she makes her own sentence: “The puddle becomes smaller as water evaporates.” Her learning action is not simply playing the audio. She retrieves the word, connects it to an observed example and checks her meaning.
Imran needs to navigate the page without relying on its visual layout
Imran uses a screen reader with an accessible worksheet. He navigates from the heading to the question and then to the answer field. If a labelled diagram has helpful text alternatives, the screen reader may communicate them; if the diagram is an unlabelled image, the technology cannot reliably supply missing information. The teacher should provide an accessible description rather than assuming the software will invent a correct one. W3C’s guidance emphasises the importance of properly structured headings, meaningful links and text alternatives.
Jia has an idea but wants help entering it
Jia says, “Evaporation happens when liquid water becomes water vapour.” Speech-to-text types a draft. Before submission she checks whether the app wrote evaporation rather than evaluation, and whether the sentence ends with a full stop. Her achievement is explaining the Science accurately; the dictation tool does not prove her spelling or punctuation is correct. In some classes, spelling and independent writing are the assessed skills, so the permitted support must follow the teacher’s instructions.
The hidden vocabulary lesson: input, output and navigation
The words input and output are useful beyond computing. Input is what enters a process; output is what comes out. In this context, typed text becomes spoken audio in TTS. Spoken audio becomes typed text in dictation. A screen reader also supplies navigation information about the interface. Students who can name the input and output are less likely to select an appealing button merely because it contains the word “speech”.
- Read aloud is an everyday instruction; text-to-speech names the conversion of text into audio.
- Dictate is a verb; dictation is the act or product. Transcribe means produce a written record of speech, while transcription is its noun.
- Navigate means move through an information structure; navigation includes headings, controls, lists and links, not merely scrolling.
- Accuracy is closeness to the intended words or meaning. A recogniser may make an inaccurate transcript even if the speaker explains the subject correctly.
- Accessible should be tested against a learner’s needs. A product claiming accessibility does not guarantee that every PDF, website or embedded diagram works properly.
Why “it can read the page” is not a sufficient test
Imagine a Science worksheet that contains a numbered heading, an explanatory paragraph, a bar chart and an input box. A basic read-aloud function may speak the paragraph beautifully but offer no practical way to reach the input box. A screen reader with correctly labelled content may announce the heading level and the form label. Neither system can repair an inaccurate chart explanation automatically. The author must give the chart sufficient accessible meaning. For the related job of watching lessons, eduKateSG already has a separate guide to captions, subtitles and transcripts; captions are not interchangeable with a screen reader.
Five diagnostic questions before changing a study routine
- Is the barrier seeing or navigating content, processing written text, or entering an answer? One student may have more than one need.
- Does the learner need spoken reading, keyboard navigation, dictation or a combination? A menu label is not an assessment.
- Can the student find the heading, diagram description, question and answer field without guessing?
- After dictation, can the student check technical terms, numbers, units and punctuation against the task?
- Is the technology permitted in this particular lesson or assessment? Check the current school and examination arrangements rather than assuming classroom preferences apply in every examination.
A 15-minute practice routine families and tutors can use
Minutes 1–3: Name the job. Give a learner three cards: “I want to hear the page”, “I want to find the next field”, and “I want to speak my answer”. Ask the learner to match each with TTS, screen reader or dictation, then explain why.
Minutes 4–7: Learn four Science words. Read an age-appropriate paragraph containing evaporate, evaporation, vapour and condense. Use the chosen access method, pause at each word and write a plain-language meaning. Note that evaporate is the verb and evaporation the noun; water vapour is a common collocation.
Minutes 8–11: Produce, do not copy. Hide the source paragraph. The student explains an example from a wet pavement, a drying cloth or condensation on a cool surface. A spoken answer can later be written or typed where appropriate. If using dictation, inspect the output for an incorrect homophone or misheard technical term.
Minutes 12–15: Transfer. Change to a Mathematics problem with symbols, or an English paragraph with quotation marks. Ask whether the tool represents expressions, units, speaker changes and punctuation accurately. Decide what still needs human checking. This is the moment when vocabulary becomes independent judgement.
From primary school to secondary pathways
Primary learners can begin with the simple contrast listen, navigate, dictate. Secondary learners may compare tools in a digital research task, discuss privacy when speech is uploaded to a service, and explain why technical spellings need verification. The distinction helps with English comprehension, Science explanations, Mathematics notation and later study or work communication. It does not guarantee examination results.
For Singapore families, avoid assuming that software features are automatically allowed in national assessments. The Singapore Examinations and Assessment Board’s SEC information explains the G1, G2 and G3 examination framework from 2027; individual access arrangements should be checked through current official channels. For wider language teaching rather than tool selection, see eduKateSG’s canonical How English Works: Vocabulary and Meaning.
Common misconceptions worth correcting early
- “A screen reader is just a louder read-aloud button.” No: navigating structured interfaces is a defining part of its work.
- “Dictation means the answer is correct.” No: recognition can mishear subject words, and the content may still be factually wrong.
- “Listening is never reading.” Listening can support access to and understanding of text, though the skills being taught or assessed determine what additional reading practice is needed.
- “Every learner must use one tool.” No: needs, preferences and devices vary; combinations can help.
- “If a PDF looks fine, it is accessible.” No: appearance alone does not tell us whether headings, reading order and diagram alternatives are usable.
Quick retrieval challenge
Without looking back, answer: (1) Which tool converts your spoken explanation into draft text? (2) Which one is designed to announce interface controls and headings? (3) Which one generally converts existing text into audio? (4) Why must a transcribed Science explanation be checked? Answers: dictation; screen reader; text-to-speech; because recognition and content accuracy require verification. For an extension, explain the difference between recognising words and recognising a speaker.
Frequently asked questions
Should every child use text-to-speech?
No. Try it where it removes a real barrier or supports the teaching goal, while continuing age-appropriate print, spelling, fluency and comprehension practice. There is no universal setting.
Can screen readers explain an inaccessible Science diagram?
Not reliably. The source must contain a good description or the teacher must provide an accessible alternative. The important language is in the diagram and its labels, not only the software.
Is dictation cheating?
That depends on the purpose and rules of a particular activity or assessment. It may be appropriate for composing ideas in one task and inappropriate where unaided spelling or written transcription is explicitly being assessed.
What should parents check first?
Ask the child to demonstrate the learning task, name the barrier and compare the input and output of a tool. Keep personal information out of trial dictation and check any account permissions or cloud processing.
The lasting skill
Children who know the words for their tools can ask for the right help, test whether it worked and switch strategies without confusion. That is a useful 21st-century vocabulary outcome: not technology for its own sake, but a learner who can read, explain, verify and communicate more independently. An accessible lesson is one in which the learner can understand and act on the meaning.
Sources and further reading: W3C WAI: Text to Speech; W3C WAI: Speech Recognition; Microsoft: Use a screen reader to dictate in Word. Device menus and access arrangements may change; consult the current provider and school guidance.