JA EN
Learning Packs
Listen to your own data. Audio learning app Sokqa

Create your own quiz instantly with AI!

By simply asking AI like ChatGPT or Claude to "Create a quiz about...", you can generate learning packs ready to be imported into Sokqa.
Follow these 3 easy steps to create your personal learning content!

  1. Send the prompt (instruction text) to an AI.
  2. Copy the generated text (JSON format).
  3. Import it using "Paste JSON" in the Sokqa app.

πŸ”° [Beginners] Let's Make One First!

Start by using the simple "Standard Prompt". It has no extra settings, making it easy for anyone to create content without errors.

1. Create 4-Choice Quizzes

Use this prompt to create 4-choice questions for vocabulary, exams, or trivia. Copy and paste the text below into an AI. 

Please create 5 multiple-choice questions on the following theme and output them in the specified JSON format.

Theme: [β˜… Enter the theme you want to learn (e.g., Frequently used Spanish words) β˜…]
Language: en

【Format to strictly follow】
{
  "id": "β˜… Any alphanumeric string (e.g., pack_1) β˜…",
  "type": "quiz",
  "schemaVersion": 1,
  "title": "β˜… Title of the learning pack β˜…",
  "language": "en",
  "questions": [
    {
      "id": "q1",
      "question": "β˜… Question text β˜…",
      "choices": ["Choice 1", "Choice 2", "Choice 3", "Choice 4"],
      "answerIndex": 0,
      "explanation": "β˜… Explanation text β˜…",
      "tags": ["β˜… Tag 1 β˜…", "β˜… Tag 2 β˜…"]
    }
  ]
}

【Rules】
- Specify 'tags' as an array of strings, setting about 1 to 3 keywords related to each question.
- The 'choices' array must contain exactly 4 items.
- The 'answerIndex' must be a number from 0 to 3 (0 represents the first choice).
- Assign a unique alphanumeric string to each question's 'id' (e.g., "q1", "q2", "q3"...).
- The output must be pure text starting with { and ending with }. Never use markdown formatting (such as ```json).
- To prevent JSON syntax errors, if you use double quotes (") within a string, always escape them (\").
- Do not use line breaks (\n) in the middle of a string. Even if the explanation is long, output it in a single line.

2. Document Mode (Reading aloud)

Use this prompt when you want the app to automatically read aloud long texts, speeches, or news articles. 

Please convert the following theme/text into Sokqa's document mode JSON format.
Split the text into individual sentences or logical phrases within the "documents" array.

Theme: [β˜… Enter the text or theme here (e.g., Steve Jobs' commencement speech) β˜…]
Language: en

【Format to strictly follow】
{
  "id": "β˜… Any alphanumeric string (e.g., doc_pack_1) β˜…",
  "type": "document",
  "schemaVersion": 1,
  "title": "β˜… Enter the title β˜…",
  "language": "en",
  "documents": [
    {
      "id": "doc1",
      "text": "β˜… 1st sentence β˜…",
      "tags": ["β˜… Tag 1 β˜…", "β˜… Tag 2 β˜…"]
    },
    {
      "id": "doc2",
      "text": "β˜… 2nd sentence β˜…",
      "tags": ["β˜… Tag 1 β˜…", "β˜… Tag 2 β˜…"]
    }
  ]
}

【Rules】
- Specify 'tags' as an array of strings, setting about 1 to 3 keywords related to the sentence.
- The output must be pure text starting with { and ending with }. Never use markdown formatting (such as ```json).
- To prevent JSON syntax errors, if you use double quotes (") within a string, always escape them (\").
- Do not use line breaks (\n) in the middle of a string. Even if the text is long, output it in a single line.

Importing the JSON into the App

Follow these steps to load the AI-generated text into the Sokqa app.

  1. Copy the entire text generated by the AI (from the opening { to the closing }).
  2. Open the Sokqa app, tap the "+" button or import button on the pack list screen, and select "Paste JSON".
  3. Paste the copied text and tap "Import".
JSON import steps in Sokqa

✨ [Advanced] Create High-Quality Audio Content

Advanced customization methods to master Sokqa's core feature: "Hands-free audio learning".

What is TTS (Text-to-Speech) Optimization?

In Sokqa, you can set a dedicated "tts" object separately from the display text.
If the smartphone's Text-to-Speech (TTS) mispronounces words or speaks too fast, adjusting the JSON text as follows will result in a much more natural, announcer-like pronunciation.

  • Create pauses with commas (,):
    If a sentence is too long, adding extra commas (,) will create natural pauses.
    Example: Listen without looking at the screen β†’ Listen without looking, at the screen
  • Add spaces to acronyms:
    If "AI" is read as a single word (like "eye"), insert a half-width space between the letters.
    Example: ChatGPT β†’ C h a t G P T, AI β†’ A I
  • Spell out special terms:
    For words that are easily misread, spell them out exactly as they sound to ensure correct pronunciation.

TTS Optimized High-Quality Prompts

These prompts instruct the AI to apply the optimization rules above, generating high-quality audio data from the start.

β–Ά For 4-Choice Quizzes

Please create 5 multiple-choice questions on the following theme and output them in the specified JSON format, optimized for Sokqa's Text-to-Speech (TTS) engine.

Theme: [β˜… Enter the theme you want to learn (e.g., SAT Vocabulary) β˜…]
Language: en

【Format to strictly follow】
{
  "id": "β˜… Any alphanumeric string (e.g., pack_1) β˜…",
  "type": "quiz",
  "schemaVersion": 1,
  "title": "β˜… Title of the learning pack β˜…",
  "language": "en",
  "questions": [
    {
      "id": "q1",
      "question": "β˜… Question text for display β˜…",
      "choices": ["Choice 1", "Choice 2", "Choice 3", "Choice 4"],
      "answerIndex": 0,
      "explanation": "β˜… Explanation text for display β˜…",
      "tts": {
        "questionText": "β˜… Question text for TTS (add more commas for natural pauses) β˜…",
        "choicesText": "Number 1, β˜… Choice 1 β˜…. Number 2, β˜… Choice 2 β˜…. Number 3, β˜… Choice 3 β˜…. Number 4, β˜… Choice 4 β˜….",
        "answerText": "The correct answer is number 1, β˜… Correct choice β˜….",
        "explanationText": "β˜… Explanation text for TTS (use spoken language) β˜…"
      },
      "tags": ["β˜… Tag 1 β˜…", "β˜… Tag 2 β˜…"]
    }
  ]
}

【Rules for TTS Object】
- Specify 'tags' as an array of strings, setting about 1 to 3 keywords related to each question.
- Adjust the text in the "tts" object to sound natural when read by a smartphone's Text-to-Speech engine.
- Add more commas (,) to create intentional pauses for better listening comprehension.
- When reading acronyms or abbreviations letter by letter, insert a half-width space between each letter (e.g., A I, U S A, P R).
- Remove unnecessary conjunctions and adjust sentences into a conversational tone that is easy to understand by ear alone.
- Follow the standard JSON rules for other fields (e.g., exactly 4 choices, output only the JSON object starting with {).

β–Ά For Document Mode

Please convert the following theme/text into Sokqa's document mode JSON format, optimized for Text-to-Speech (TTS).
Split the text into individual sentences or logical phrases within the "documents" array.

Theme: [β˜… Enter the text or theme here (e.g., Steve Jobs' commencement speech) β˜…]
Language: en

【Format to strictly follow】
{
  "id": "β˜… Any alphanumeric string (e.g., doc_pack_1) β˜…",
  "type": "document",
  "schemaVersion": 1,
  "title": "β˜… Enter the title β˜…",
  "language": "en",
  "documents": [
    {
      "id": "doc1",
      "text": "β˜… 1st sentence for display β˜…",
      "tts": {
        "text": "β˜… 1st sentence for TTS (add more commas for natural pauses) β˜…"
      },
      "tags": ["β˜… Tag 1 β˜…", "β˜… Tag 2 β˜…"]
    },
    {
      "id": "doc2",
      "text": "β˜… 2nd sentence for display β˜…",
      "tts": {
        "text": "β˜… 2nd sentence for TTS β˜…"
      },
      "tags": ["β˜… Tag 1 β˜…", "β˜… Tag 2 β˜…"]
    }
  ]
}

【Rules for TTS Object】
- Specify 'tags' as an array of strings, setting about 1 to 3 keywords related to the sentence.
- Adjust the text in the "tts" object to sound natural when read by a smartphone's Text-to-Speech engine.
- Add more commas (,) to create intentional pauses for better listening comprehension.
- When reading acronyms or abbreviations letter by letter, insert a half-width space between each letter (e.g., A I, U S A, P R).
- Remove unnecessary conjunctions and adjust sentences into a conversational tone that is easy to understand by ear alone.
- The output must be pure text starting with { and ending with }. Never use markdown formatting (such as ```json).

Troubleshooting & Manual Fixes

If you encounter an error during import, check for the following common causes.

Common Errors

  • The choices array has 3 or 5 items β†’ It must have exactly 4.
  • answerIndex is set to 4 β†’ If the 4th choice is correct, specify 3.
  • There is a trailing comma ( , ) after the last item β†’ Remove the extra comma (,) immediately preceding a closing brace } or bracket ].

Sokqa JSON Format Specification

This explains the structure of the JSON data imported into Sokqa and the meaning of each tag.
You can run the app with just the bare minimum "Required Tags", or add "Optional Tags" for a richer experience.

Required Tags (Bare minimum to run)

Tags that are absolutely necessary for the app to recognize and run the learning pack.

  • schemaVersion : Always specify 1.
  • type : Specifies the learning format (quiz = 4-choice, document = reading aloud).
  • id : A unique ID assigned to the entire pack or each question/document (e.g., pack_1, q1). Must be alphanumeric and unique.
  • title : The title of the learning pack. Displayed on the app's list screen.
  • language : The language code (e.g., en for English, ja for Japanese).

[Required for 'quiz' (4-choice)]

  • questions : A list (array) of questions.
  • question : The question text (displayed on the screen).
  • choices : The choices. You must provide exactly 4 choices.
  • answerIndex : The correct answer number. Specify a number from 0 to 3 (0 is the 1st choice, 3 is the 4th).

[Required for 'document' (Reading aloud)]

  • documents : A list (array) of sentences. Split by meaning or one sentence at a time.
  • text : The text of the sentence (displayed on the screen).

Optional Tags (For a higher quality experience)

Not required, but adding these allows you to fully utilize the app's features.

  • explanation (quiz only) : Explanation text for the question. Displayed on screen when answering correctly/incorrectly.
  • tts (quiz / document) : An object specifying text exclusively for Text-to-Speech (TTS). Use this to set conversational tones, explicit spellings, or pauses (commas) separately from the display text.

    [tts structure for 'quiz']

    • questionText: Question text for reading aloud
    • choicesText: Choices text for reading aloud (e.g., "Number 1, XX. Number 2, XX.")
    • answerText: Correct answer text for reading aloud (e.g., "The correct answer is number 1, XX.")
    • explanationText: Explanation text for reading aloud

    [tts structure for 'document']

    • text: Document text for reading aloud
  • tags (quiz / document) : A list (array) of tags to attach to the question or sentence. This can be used for tag search and filtering learning targets within the app (e.g., ["Vocabulary", "TOEIC"]).
Back to Top