PromptGarden

Author: Prompt Expert

  • The Memory Keeper: Turn a Family Recording into a Searchable Archive

    The Memory Keeper: Turn a Family Recording into a Searchable Archive

    There is an old family video on your computer, or a recording of a relative telling a story. You would like to read what was said and find the important moments again without replaying the whole thing.

    This prompt turns that recording into a transcript and a short description you can keep beside it. October 27, World Day for Audiovisual Heritage, is a good occasion to start with one.

    Start with the file you already have

    For example, open Gemini, choose Add files → Upload Files, and select your audio recording or video. Once it is attached, paste the prompt below. You can use another assistant, including ChatGPT, if it supports transcribing the file you upload.

    For a walkthrough, see how to upload an audio or video file and get a transcript. You do not need to type the spoken words yourself.

    The prompt

    Help me preserve this family recording.

First, transcribe the spoken words in the attached audio or video file. Keep the original language and wording. Mark anything you cannot hear clearly as [unclear]. Use neutral speaker labels when you can distinguish the voices. Do not guess names or missing words.

Show me the transcript and wait for me to check it.

After I confirm it, add a short description with:
- a clear title;
- a summary of no more than 100 words;
- people, places, and dates actually mentioned;
- five keywords that will help me find it later;
- any unanswered questions worth asking my family.

Keep uncertain memories uncertain. Distinguish the date of the recording from dates mentioned in the story.

Return the checked transcript and the description as two clearly labelled sections I can copy into a document. If you cannot read the attached recording, tell me.

    Check the memories, then save them

    Listen again where the transcript is uncertain. Correct a misheard name, but keep phrases such as “after the winter, I think.” The aim is to preserve what someone said, including the uncertainty, rather than turn it into a more definite story.

    Copy the finished transcript and description into Word, Google Docs, or another document. A name such as Grandma’s first job — transcript and notes makes the result easy to recognize. Download the document if you want to keep a local copy.

    Where your family archive lives

    Create a folder called Family Recordings on your computer or in a cloud drive you use. Make a subfolder called Grandma’s first job and put two things inside: the original recording and your new document. That is your first archive entry. Repeat for the next recording.

    The assistant’s answer stays in the chat until you copy or download it; you choose where the saved files go. Give folders descriptive names, keep a second copy elsewhere, and decide who can access a shared folder. A useful family archive can begin with just one recording and one readable document.

  • The Brain Dump Prompt: Turn Messy Notes into Three Clear Next Steps

    The Brain Dump Prompt: Turn Messy Notes into Three Clear Next Steps

    Your notes contain a reply you owe someone, an idea for next month, a half-finished task, and a reminder you have copied three times. Before you can get started, you have to work out what actually needs doing.

    This prompt turns that jumble into up to three clear next steps, with the remaining items kept in a short list for later. It is a useful way to begin a week or restart a day that has become scattered.

    Start with the notes you already have

    Open a new chat in your AI assistant. Copy in your rough notes from your phone, a document, or a few lines you type on the spot. Leave them messy. Add your main goal and the time you have available, then send the prompt below.

    The prompt

    Turn these messy notes into a manageable action list.

My main goal: [GOAL]
Time available today: [TIME]
My notes: [PASTE YOUR NOTES]

Separate tasks, ideas, and unanswered questions. Combine duplicates without losing important details.

Suggest up to three next actions that fit my goal and available time. Make each action specific enough that I know how to start. Briefly explain why you chose it.

Keep everything else in a short “Later” list. Do not invent deadlines or treat every idea as a commitment. Flag missing information instead of guessing.

    What a useful answer looks like

    Imagine your goal is to prepare Friday’s newsletter and you have thirty minutes today. Your notes mention a missing quote from Maya, an unfinished opening, a registration link to check, a possible podcast, and a website redesign.

    An illustrative shortlist might be:

    1. Ask Maya for the missing quote. This starts a task that depends on someone else.
    2. Draft the newsletter’s opening paragraph. You can make progress while waiting.
    3. Open the registration link and check where it leads. This resolves one concrete uncertainty.

    The podcast and redesign belong on the “Later” list unless you decide otherwise. That distinction matters: writing down an idea should not automatically turn it into another obligation.

    Take the list out of the chat

    Check the suggestions against your real commitments. The assistant cannot know an unmentioned deadline or how long a particular task takes you. If an action is too large, ask: “Reduce this to a first step I can finish in ten minutes.”

    Copy the actions you accept into your usual notes or task app and put the “Later” items underneath or on a separate page. The result initially exists only in the chat; copying it makes it part of the list you actually use. Start with the first action, and revisit the rest when your time or priorities change.

  • From Voice to Text: Create a Usable Transcript with ChatGPT and other LLMs

    From Voice to Text: Create a Usable Transcript with ChatGPT and other LLMs

    You have an interview on your phone, an old family video, or a voice memo you want as text. An AI assistant can turn the spoken words into a transcript. You upload the existing file and ask it to write down what is said.

    Upload your recording

    Here is a straightforward example using Gemini, which supports both audio and video uploads:

    1. Open Gemini and sign in.
    2. Choose Add files → Upload Files beside the message box.
    3. Select the recording already saved on your device and wait for it to attach.
    4. Paste the prompt below and send it.

    Start with a short clip. For a video, use the video file itself; the prompt asks for its spoken words. Google’s upload guide explains the current file and length limits. If your recording is too long, try a shorter section first.

    You can also use this prompt in ChatGPT or another assistant when it supports transcription of your uploaded file. If your current chat cannot read the recording, use the Gemini steps above.

    The prompt

    Please transcribe all spoken words in the attached audio or video file.

Keep the original language and wording. Add punctuation and paragraphs so the text is easy to read. Do not summarize or translate it.

Use Speaker 1, Speaker 2, and so on when you can distinguish different voices. Write [unclear] wherever you cannot make out a word instead of guessing. Preserve corrections such as "1982—no, 1983."

For a video, transcribe the speech without adding descriptions of the pictures.

Return the transcript, followed by a short list of passages I should listen to again. If you cannot access the recording, tell me instead of inventing its contents.

    Check it and save it

    The result appears in the chat. Play back a few passages while reading along, especially names, dates, numbers, and anything marked [unclear]. Check the beginning and end too: a fluent answer is not proof that the whole recording was covered.

    Make your corrections, then copy the text into Word, Google Docs, or a plain text file. Give it the same name as the recording, such as grandmas-first-job-transcript.docx, and save it beside the original. If the assistant offers a downloadable document, download that instead.

    Keep the audio or video as well: the transcript captures words, while the recording preserves the voice. Upload recordings only when you have permission to share them with the service. For a family collection, The Memory Keeper adds a short description that helps you find each recording again.

  • Tiny Worlds: Turn an Everyday Object into a Miniature Landscape

    Tiny Worlds: Turn an Everyday Object into a Miniature Landscape

    A cup can hold a valley. A worn shoe can become a hillside. The interesting part is keeping the ordinary object recognizable while giving it a new scale and a small world of its own.

    This image prompt uses a reference photograph as the structure for an imagined landscape. Choose an image tool that accepts reference photos, then decide which parts of the object must remain visible.

    The prompt

    Use the attached photograph of [OBJECT] as the visual reference for a miniature landscape.

Keep the object's main silhouette, recognizable proportions, material, and defining feature [HANDLE / LACES / LENS / OTHER FEATURE]. Keep it as the single central subject.

Transform only [THE OPENING / INTERIOR / SELECTED SURFACE] into [LANDSCAPE]. Make the landscape feel physically integrated with the object through coherent perspective, lighting, and contact shadows.

Include one clear scale cue: [A TINY PATH / SMALL TREES / A LITTLE BRIDGE]. Limit secondary details so the object remains easy to recognize at a glance.

Style: [WATERCOLOR ILLUSTRATION / PHOTOREALISTIC DIORAMA / PAPER SCULPTURE].
Lighting: [SOFT DAYLIGHT / WARM EVENING LIGHT].
Background: simple and quiet, with generous empty space around the subject.
Composition: square, with the complete object visible.

No added text, logos, watermarks, extra copies of the object, or unrelated scenery.

    Choose an object with a clear shape

    A simple reference makes the transformation easier to describe. Photograph the object against a quiet background with its defining feature visible. A cup shown from slightly above gives the model both a recognizable handle and an opening where a landscape can sit.

    For a first attempt, try an ivory teacup containing pale green hills, a narrow footpath, and three tiny trees. Keep the cup’s rim and handle intact. The footpath supplies scale; the trees establish the world without requiring a miniature town, waterfall, railway, and crowd of people.

    The medium changes the character of the result. Watercolor can make the transition between ceramic and hillside feel gentle. A photographic diorama asks for more convincing material and lighting relationships. Paper sculpture turns the scene into a visibly constructed object. Choose one before adding more stylistic instructions.

    Revise one relationship at a time

    Inspect the object first. If its handle has vanished or its outline has become an ordinary mountain, ask for a revision that restores that feature while preserving the existing landscape. If the scene is crowded, reduce the number of elements rather than requesting more detail everywhere.

    Next, check where the two worlds meet. Do the tiny trees appear rooted in the landscape? Does the landscape sit inside the cup or float above it? A focused follow-up might be: “Keep the composition and palette. Lower the hills inside the rim and add a soft contact shadow where they meet the ceramic.”

    A reference image guides generation; it does not guarantee exact preservation. Check the result if the object has personal or commercial significance. For a series, keep the background, lighting, and medium consistent, then change just the object and its landscape. Each image can have its own small surprise while still belonging to the same collection.