Are my images or videos uploaded to MyToolyard?

No. Quick text runs entirely in your browser. In Smart AI mode, images and sampled frames go directly from your browser to the AI provider whose key you added — never through our servers.

How does it read a video?

It takes still frames at the interval you choose (or whenever the scene changes), reads each one, removes repeats and merges the results into one entry per video with timestamps.

How many files can I process?

There is no fixed limit. Large queues run in batches you choose; progress is saved in this browser so you can continue later.

What does the Excel file look like?

The first sheet has one row per file with columns for type, confidence, summary, each extracted field and the full text. Each file also gets its own sheet with its fields and text line by line.

Which mode should I use?

Quick text is ideal for clear printed text. Choose Smart AI for handwriting, tables, forms, charts, photos of real-world scenes or anything where you want specific fields.

What happens if my AI provider rate-limits me?

The tool waits and retries automatically. If your key is rejected, the batch pauses and the rest of your queue stays safe.

Image & Video to Text Extractor

Drop in hundreds of photos, scans, screenshots or videos, pull out text and data file by file, then download it as Excel, Word, PDF, CSV or JSON.

No training on your data
Region: global

Media Studio

How to read files

Text languageVideo frames

Media StudioAI Extraction & Knowledge

Image & Video to Text Extractor — Overview

About this tool

What is Image & Video to Text Extractor?

The Image & Video to Text Extractor turns a pile of photos, scans, screenshots and video clips into organised, searchable information. Every file gets its own result card, so the text from slide 12 never blurs into slide 13 — and when you download, each source keeps its own row in Excel, its own section in Word and its own page in PDF.

There are two ways to read files. Quick text runs a text-recognition engine inside your browser: it's free, it needs no account and the images never leave your device. Smart AI sends each image or sampled video frame straight from your browser to your own AI provider (Gemini, OpenAI, Claude, OpenRouter or Groq), which can understand tables, forms, receipts, charts and what is happening in a scene.

It's built for big jobs. Add hundreds of files, choose a batch size, and process them in rounds. The queue is saved in your browser, so you can close the tab, come back tomorrow and press Continue — finished files stay finished and failed ones can be retried.

What it does

What does Image & Video to Text Extractor do?

Everything you need to get information out of media in bulk:

Images (JPG, PNG, WebP, HEIC) and videos (MP4, MOV, WebM) in one queue
Video frames sampled every 2, 5 or 10 seconds, or at scene changes, with timestamps
AI goals: full transcription, a table of fields, a summary or your own instructions
Per-file cards with confidence scores, editing, copy and re-run
Excel with one row per file plus a sheet per file; Word, PDF, CSV, JSON and TXT
Resumable batches with pause, retry and automatic waiting on rate limits
Send results to Knowledge Agents to chat with them later

How to use

How to use Image & Video to Text Extractor

1
Add your files
Drop, paste or browse for images and videos. They're stored in this browser so a large queue survives a reload.
2
Pick the engine and batch size
Quick text for plain printed text at no cost; Smart AI for tables, forms and video understanding using your own key. Set how many files to process per round.
3
Run, review and download
Press Start. Check each card, fix anything by hand, then download all or selected files as Excel, Word, PDF or more.

Good to know

Privacy & cost

Your API key stays in your browser — requests go directly to your chosen provider. Usage is billed to your account, not ours, so you stay in full control of model choice and spend.

Use cases & scenarios

Where it fits

Invoices and receipts

Use 'Table of fields' to turn a folder of receipts into a spreadsheet of vendors, dates and totals.

Lecture slides and whiteboards

Photograph every slide, extract them in order and download one Word document with a section per slide.

Screen recordings and tutorials

Sample a video at scene changes to capture every on-screen step with a timestamp.

Archiving paper records

Process hundreds of scanned pages over several sessions without losing your place.

Value & benefits

Why people pick this tool

✓Quick text is free and fully on-device
✓Smart AI uses your own key — we never see your files or your key
✓Results stay separate per source, on screen and in every export
✓No file-count limit, resumable over days

AI integration & provider setup

Bring your own key (BYOK)

Image & Video to Text Extractor uses your own AI provider via Bring-Your-Own-Key. Your key is stored only in this browser's localStorage and sent directly to the provider you choose — MyToolyard never sees, logs, or proxies your traffic.

GeminiOpenAIClaudeOpenRouterGroq

Open ⚙️ Settings in the top-right of the page.
Paste an API key from your chosen provider's dashboard.
Save — the key persists in this browser only. Return here and run the tool.

Open Settings

Frequently asked questions

FAQs

Final thoughts

The bottom line

Getting information out of images and video shouldn't mean retyping it. Queue everything, let it run in rounds, and download a tidy file where every source keeps its own place.