Matt Campbell on Nostr: Question for users of NVDA and web-based Mastodon clients like Semaphore: When ...
Question for users of NVDA and web-based Mastodon clients like Semaphore: When looking at a toot with an image, what's the easiest way to run high-quality OCR on the original, full-sized version of the image? Pressing NVDA+R runs OCR on the image as it's rendered on the screen, which is possibly a shrunken version, leading to less accurate OCR. Am I missing something obvious?
Published at
2024-06-24 18:32:18Event JSON
{
"id": "b4f163fe342663f10ff7b9553c478ca7b5978fb444c1d0a64d7efa223d9892d5",
"pubkey": "790e5e1a77f369f0eae0afb55dd1fae16c51eb87977c87a9b4fa113c6bac5399",
"created_at": 1719253938,
"kind": 1,
"tags": [
[
"proxy",
"https://toot.cafe/users/matt/statuses/112673026094105919",
"activitypub"
]
],
"content": "Question for users of NVDA and web-based Mastodon clients like Semaphore: When looking at a toot with an image, what's the easiest way to run high-quality OCR on the original, full-sized version of the image? Pressing NVDA+R runs OCR on the image as it's rendered on the screen, which is possibly a shrunken version, leading to less accurate OCR. Am I missing something obvious?",
"sig": "44d8e81f861cee5fa88ad5bb1645ca4d876fc621525fe729bb459b6e9e4272d68e48db4b85668e0de2576001ac86d17cb1467db8f0f443b771cfcebc171c3fd2"
}