This deck shows you British English words in regular orthography, and then you have to guess the IPA (International Phonetic Alphabet) transcription of their Received Pronunciation. The idea is that one can improve their pronunciation by drilling IPA. Please notice that this proposal is unusual, experimental, untested, and possibly wrong.
Alternatively, you can ignore or skim the IPA and use this deck just for the audio files; see if you guess how the words sound, then test whether you were right. This works better if you record yourself and compare the recording to the sample audio. Anki Deskop has a "Record Own Voice" (Shift+V) option at the "More" card menu. With or without IPA drilling, I recommend careful attention to both listening and speaking practice.
The deck includes 1000 words, ordered by frequency. Hopefully this will be enough to give you a good grounding on English pronunciation, fix any common words you might be mispronouncing, and also get a good feel for the complex relationship between writing and sound in English.
First, ensure your computer has IPA fonts. Noto Sans/Serif are free and good enough.
Read or skim the Wikipedia article "English phonology" (unless you're already familiar with the topic). In particular, make sure you're aware of the following points:
- Schwas (/ə/), and how any orthographic vowel may be read as a schwa in unaccented syllables.
- Dark L.
- Non-rhotic accents (including British Received Pronunciation) and the "Linking R".
- Aspiration of voiceless consonants at the start of words.
- Assimilation and glottalization of consonants.
- /ʌ/ is actually [ɐ], /ɒ/ is actually a short [ɔ], and /e/ is actually [ɜ]. English transcriptions commonly use the wrong IPA symbols because of "tradition". Yes, I'm baffled too.
This will help prevent confusion between apparent transcription mismatches. If you have doubts about some sounds, BBC has an excellent series of short pronunciation videos, focusing on one sound at a time (they're also available on Youtube). Watching them all at least once is advisable.
Then browse the article on "English orthography". Make sure you understand the roles of the "silent E", double consonants, etymological letters, and morphophonemic (or "underlying") representation.
Ensure your deck options are set for Show new cards in order added. This will optimize your time, since the most frequent cards are listed first.
Finally, start your flashcarding! Read the word in the question side and write the IPA transcription somewhere (the most convenient method may be a paper notepad by your computer, or Ankidroid's whiteboard function). See if the answer matches, and also listen to the audio carefully (press F5 to repeat). Make sure you got all consonants and vowels right; and, furthermore, take notice of:
- Silent letters.
- Short vs. long vowels.
- Neutralization (which vowels become schwas).
- Position of stress. Because of the tools I've used, the transcription puts the stress mark (ˈ) before the stressed vowel. Most sources put it before the consonant, marking the entire syllable. But this is just a convention and doesn't really matter; just make sure you noted down which syllable is stressed.
Both the IPA transcriptions and the audio are automatically generated. Both may have mistakes. Trust your ears; if they differ, the transcription is more likely to be wrong than the audio. If you doubt, you can suspend a card; or else go check a source (Wiktionary, OALD) and fix the card. Reports of issues will be welcomed.
Notice also that some English written words may be pronounced in different ways depending on context (heteronyms); for example, the word "the" is read as /ðiː/ when emphatic, but /ðə/ when fluent; and "conflict" as a verb is /kənˈflɪkt/, but as a noun it's /ˈkɒn.flɪkt/. This deck tried to avoid this kind of ambiguous orthography, but some may be left. This may generate discrepancies between the audio and transcription, and also your own transcription. When in doubt, check a dictionary.
Schmidt's Noticing Hypothesis postulates that second-language (L2) learners may never acquire a phoneme or other feature unless they're first brought to attention to it. For example, a learner may spend years pronouncing /θ/ as /s/, no matter how much auditory input they get; the input only starts to matter after they learn about the existence of /θ/, how it's produced, how it sounds etc.
The Noticing Hypothesis matches my own experience, and also that of many other language teachers/learners. It's not, however, consensus in linguistics, nor is it experimentally validated. It may be wrong.
This deck is based on an even stronger (or wonkier) version of the Noticing Hypothesis. Call it the Acoustic Image Hacking Conjecture (AIHC). An "acoustic image" is the mental model one has of a word's sound; or, more accurately, it's the word's symbolic representation, in your mind, as a string of abstract phonemes. The AIHC posits that, if you have internalized the wrong acoustic images, it may be possible to change them directly by conscious reexamination of their phonological sequence; and, furthermore, that written IPA is a good tool for this reexamination. The AIHC does not fit very well with many widespread assumptions in linguistics about the primacy of speech over writing, and about the efficacy of conscious metalinguistic analysis vs. natural acquisition. It may well be complete bunk. Of course, I personally think it's likely to be an effective technique; I wrote this deck for my own practice, after all.
If you try to use this deck to improve your pronunciation and it works, or fails to, I'd be very interested in hearing about your experience.
To generate this deck I've used Espeak (IPA transcriptions), Google Translate through Awesome TTS (audio synthesis), Sublex-UK and wordlist.sourceforge (word frequency data). I've also used heteronym lists by John Higgins, Jon Vahsholtz and Wiktionary, with the help of The Briticiser for spelling.
Leonardo Boiko leoboiko@namakajiri.net