PhrasePump Suggestion

[CPUB0T]

@Angelina_Kalinichenk

PhrasePump is a great tool for becoming more familiar with the vocabulary you encounter through immersion. It strengthens memory because it demands active recall, not just recognition. That said, PhrasePump has a few weaknesses that could be addressed with relatively small updates.

Ubove is an example card from my immersion that illustrates what I think would make PhrasePump more effective.

Right now, the biggest issue is that translation is visible immediately. That becomes an unavoidable crutch for many learners: your eyes naturally jump to what’s familiar, and you end up “confirming meaning” instead of engaging with the target language first.

I’ve already revealed the target word in this example so I can point out something else: the popup dictionary provides three short, one-word definitions. Those definitions would work far better as progressive hints than showing the full translation and/or full sentence context from the start. Even those hints should be hidden initially so the learner gets a clean opportunity to recall. At this point, we do not want transliterations. We must focus on recall.

Proposed workflow for PhrasePump cloze cards

1. Start with the sentence clozed (target word hidden).
   Keep translation hidden and hints hidden.

• This forces the learner to interact with the target language first.
• It prevents “translation-first scanning,” which undermines recall.

2. Reveal the target word next.
   This step is especially helpful for Asian languages (or any language with a writing system very different from the learner’s native language).

• It tests reading directly.
• It supports gradual acquisition: you’re not only learning meaning, you’re building visual familiarity with the word’s form.
  It may be worth giving the end user some options here. If the target language uses a writing system similar to the learner’s native language, a visual reveal can become too easy and turn into simple recognition rather than recall.

3. Reveal hints after that.
   Show the short dictionary hints (like the one-word definitions) before the full translation.

• This gives the learner a chance to retrieve meaning with a light prompt instead of jumping straight to the answer.
• It’s a middle step between “no help” and “full solution,” which is ideal for memory formation.

4. Reveal the full translation next.
   Only after the learner has attempted recall (and possibly used hints) should the full translation appear.

• This provides confirmation and deeper context.
• It also helps resolve ambiguity when a word has multiple senses.

5. Play the audio last and integrate everything.
   At this point, the learner can connect: spelling/form → meaning → full sentence meaning → pronunciation/intonation.

• This is where the card becomes a “complete rep” that ties together reading, meaning, and sound.

Alternative use case: listening-first mode

If the learner’s main goal is listening, the flow should start with audio first, while keeping everything else the same:

• Play audio first (no text revealed yet).
• Then proceed with the same reveal order: clozed sentence → target word → hints → translation.

This preserves the same core principle (avoid translation-first dependency) while training the ear before the eyes.

Everything can stay the same, with two changes:

• Hide the translation by default so it isn’t visible immediately.
• Add a “Reveal Hint” button (short dictionary-style definitions). Buttons or hover reveal allows the user to choose which feature they want to use in what order.

Audio stays optional and user-controlled, just like it is now.
Also could have an Audio focused mode where everything is blurred, even the target language, very much like PhrasePump’s Practice Mode.

Edit: user notes on a card would also be nice. We could add our own hits, definitions, explinations and things.

[Angelina_Kalinichenk]

Hello @CPUB0T,
Thank you for such a detailed and well-thought-out plan for improving PhrasePump. You have proposed a really strong solution.
As for the fact that the translation is now immediately visible: a similar approach has already been implemented in PhrasePump’s Listening Practice mode. When you switch to a card, the audio is played automatically first, with both the subtitles and the translation hidden. You can then reveal the information gradually:
– first show the subtitles,
– then the translation,
using the “E” key on the keyboard, or hover over the card and reveal everything at once.

Thus, Listening Practice already uses logic similar to what you describe:
first audio → then text → then translation.
In this mode, the main emphasis is indeed on developing listening comprehension skills.

However, you are absolutely right that this logic is currently lacking in the Quiz / PhrasePump cards modes. The phrase and translation are visible immediately, and this can indeed provoke “translation-first scanning” and reduce effectiveness. Your suggestion of progressive disclosure of information (cloze → target word → hints → translation → audio) could significantly enhance the learning effect of PhrasePump and accelerate the assimilation of saved items.
I will create a ticket with your suggestions and pass them on to the team as an idea for improving PhrasePump, primarily for Quiz mode.

You can also duplicate your proposal in the Telegram channel specified in the email with the promo code. There you can engage in a more lively discussion and perhaps explore this concept in greater detail together.