Video2Text OCR+ 1.0.5

Snapshots containing visual text, formatted PDF text, and plain text can be extracted from a video. Utilizing multi-core CPU and GPU of modern hardware, along with intelligent algorithms that minimize redundant calculations across frames, the process can be completed in a fraction of the actual video duration. On-device text recognition ensures privacy protection and operates without an internet connection.

Three distinct output formats are created:

1. Multi-page PDF: Consists of snapshot images of video frames with recognized text. In cases where similar text appears across multiple frames over a period, the clearest display is selected. This format is ideal for presentation videos and includes optional annotations of the original frame time for easy navigation.
2. Text-only PDF: Corresponds to snapshot images with formatted text approximating the original layout. Options for black or white text on varying backgrounds are available. This format is significantly smaller in file size compared to snapshot images and allows for easy copying and searching of text.
3. Plain Text: Combines all pages in a plain text format with customizable formatting options. Features an embedded text editor for correcting recognized text. The file size is even smaller than the text-only PDF format.

Both the text-only PDF and plain text formats offer automatic detection of specific content such as emails, phone numbers, web links, and addresses, allowing quick access to relevant apps with a single click.

An intuitive interface guides users through the app:

1. Step 1: Various input methods provided include selecting videos from the Photos or Files app, as well as drag-and-drop functionality on iPadOS. The app can also be accessed via mail attachments or file sharing from other apps.
2. Step 2: Users can opt for default settings or customize preferences such as output formats, filename allocation, annotation of frame time in PDFs, dark mode for PDF text, and plain text formatting options.
3. Step 3: Preview snapshots while awaiting final processing. Three output formats can be reviewed after completion. Device gestures enable zooming in PDFs or adjusting font sizes.
4. Step 4: Distribute recognized text via various sharing options depending on the selected format. Functions include airdrop, messaging, cloud storage saving, social media posting, printing, or copying to compatible apps.

The app's current version prioritizes recognizing printed English text in upright positions and emphasizes stable text presence for at least a second to ensure accurate detection. Notably, flashing, scrolling, or fast-moving text may not be captured intentionally.

It offers auto-renewable subscriptions with an introductory free trial period for full access to features. Users are encouraged to review the terms of use provided at https://thingkingland.app/legal/termsOfUse.html.