Dictation

Dictation, in language learning, is the act of transcribing spoken language into written text. The dictation method has roots in classical schoolrooms — students would write down what a teacher read aloud, and the teacher would mark errors. Modern dictation does the same job, just with audio recordings and instant feedback instead of a teacher with a red pen.

The English word dictation comes from the Latin dictare, meaning to say repeatedly or to dictate. The word entered English in the 17th century and has carried the same core meaning ever since: producing written text from spoken sound. Today dictation appears in two contexts — speech-to-text dictation (you speak, the machine writes) and language-learning dictation (you listen, you write). This page is about the second kind: dictation as a listening exercise.

A dictation exercise is small but unforgiving. You hear a sentence. You type it. The system shows you which words you nailed and which you missed. There is no multiple choice to hide behind. There is no skim-and-skip. Dictation surfaces what you actually heard — and what you didn't.

When you read English, your eyes can backtrack and your brain has unlimited time. When you listen to English, the audio is gone in seconds and you have to decode it in real time. Dictation isolates exactly this real-time decoding: phoneme recognition, word segmentation, grammar parsing, and short-term memory all run at once.

Researchers in second-language acquisition call this the noticing hypothesis: you only learn a feature of a language once you consciously notice it. Dictation forces noticing. If you keep mishearing was as is, dictation will mark it wrong every single time until you start noticing the difference. Passive listening lets you smooth over the gap; dictation does not.

Dictation also trains the bottom-up listening skills that most learners under-train. Bottom-up means starting from sounds and building up to meaning, the opposite of guessing meaning from context. Most listening practice (podcasts, TV, conversation) is top-down — your brain fills gaps from context. Dictation removes context as a crutch and forces you to actually decode the sound.

There is no single dictation method. The classic full-sentence dictation is just one variant. Modern dictation tools usually offer several:

• Full-sentence dictation. You hear the whole sentence and type every word. The hardest variant. Strongest training for short-term memory and full-sentence parsing.
• Gap-fill dictation. You see most of the sentence with some words blanked. You only type the blanks. Easier; great for targeting specific words (function words, weak forms, homophones).
• Adaptive dictation. Difficulty adjusts to your level — Edictatio defaults to this. Words you've shown you can spell are pre-filled; tricky words remain blank.
• Single-word dictation. One word at a time. Best for vocabulary review and pronunciation-to-spelling mapping.
• Shadow dictation. You listen and speak along, then write. Combines dictation with shadowing — slower but trains pronunciation simultaneously.

English is a uniquely brutal language to listen to. Spelling barely tracks pronunciation (compare though, through, thought, tough). Function words like was, of, and the get reduced to weak vowels and almost vanish. Native speakers chain words together until sentence boundaries blur. Standard listening practice rarely trains any of this; dictation drills all of it.

Dictation also forces accurate spelling — a side benefit most learners undervalue. If you can't spell a word, you don't really know it. If you can spell separate, definitely, accommodate, and rhythm under dictation conditions, you've genuinely learned them. Dictation teaches you the spelling at the same time it teaches you the sound.

Finally, dictation teaches grammar implicitly. When you hear she's been working and have to type all four words, you confront the contraction, the auxiliary, the past participle, and the -ing form together. Dictation is the closest thing to a single drill that trains listening, spelling, vocabulary, and grammar at the same time.

Most English exams test listening. Most learners under-prepare for the listening section because passive practice (watching shows, listening to podcasts) feels productive but rarely closes specific gaps. Dictation closes specific gaps fast.

For IELTS Listening: dictation on accent-varied audio (British, Australian, North American) trains the accent-shift you'll face. For TOEFL Listening: dictation on academic lectures trains the specific register and pacing of the test. For Cambridge B2/C1/C2: dictation across genres builds the broad listening base those exams reward. Edictatio's topic library is organized exactly along these lines.

A common pre-exam routine: 15 minutes of focused dictation per day, alternating between exam-style audio and one off-genre topic for breadth. Dictation in short, daily doses outperforms long weekend sessions — this is the same spaced-practice principle that makes FSRS effective.

• Dictation vs passive listening. Passive listening (podcasts, shows) builds familiarity. Dictation builds accuracy. Both matter, but only dictation gives you measurable feedback on what you missed.
• Dictation vs shadowing. Shadowing trains pronunciation and rhythm. Dictation trains decoding and spelling. They are complementary, not substitutes.
• Dictation vs transcription. Transcription is dictation at scale — full audio transcribed to text. Useful but exhausting. Sentence-level dictation gives you tighter feedback loops.
• Dictation vs translation. Translation tests comprehension. Dictation tests perception. Many learners can translate fluently but still fail at dictation — the perception layer is genuinely separate.

Doing one round of dictation on a sentence and never seeing it again wastes most of the learning. The same memory science that powers Anki — spaced repetition — works for dictation too. Words you missed should come back later, at increasing intervals, until they stick.

Edictatio uses FSRS (Free Spaced Repetition Scheduler), the same modern algorithm Anki adopted. Every word you miss in a dictation session enters a personalized review schedule. Easy words come back rarely; tricky words come back often. The whole point of dictation is to stop missing words — FSRS makes sure you don't drift back.

This pairing — dictation as the assessment, FSRS as the retention engine — is the core of Edictatio. Most dictation tools stop after the session. Edictatio remembers what you missed and brings it back at the right moment.

1. Pick a topic at your level. Don't start with native-speed news if you're B1. Pick a topic where you understand 60–80% on first listen. Dictation at 0% comprehension is just frustration; dictation at 90% is too easy.
2. Listen once without typing. Get the gist. This activates context and gives your brain a frame for the sentence.
3. Dictate sentence by sentence. Play, pause, type. Replay if you need to — Edictatio lets you replay freely without penalty.
4. Submit and read the diff. Don't skim. Read every word you got wrong. Notice the pattern: did you miss a function word? a contraction? a homophone?
5. Trust the FSRS queue. Tomorrow, words you missed today will resurface. Don't fight the queue, don't skip reviews. Consistency beats intensity for dictation.

• Dictating audio that's too hard. If you're missing 50%+ of words, the audio is above your level. Drop down. Dictation is most effective in the 70–90% comprehension band.
• Replaying 20 times to hit 100%. After the third replay, you're memorizing the audio, not training your ear. Submit, take the hit, and let FSRS bring the missed words back.
• Ignoring the diff. The post-submission diff is the most valuable part of dictation. If you only check the score, you skip the actual learning.
• Practicing 2 hours once a week. Dictation rewards short daily sessions. 15 minutes a day beats 2 hours on Sunday — for both retention and motivation.
• Mixing dictation languages. Pick one target accent (or one varied target like IELTS) and stick with it for at least a few weeks. Constant accent-switching slows pattern formation.

Edictatio is built around the assumption that dictation is the most efficient listening drill that exists, but only if the feedback loop is tight. Sessions are short. The diff is per-word. The FSRS queue is automatic. There is no upsell, no ad.

The audio is real — IELTS prep, TOEFL lectures, podcasts, native-speaker recordings. No synthetic voices, because dictation against TTS is dictation against an artifact, not against actual English. Topics are organized by genre and difficulty so you can pick what matches your goal: dictation for exams, dictation for casual listening, dictation for academic English.

Edictatio is free. Free as in no ads, no upsell. The project is small and run by one person. The whole point is to give serious learners a serious dictation tool.