{"id":163,"date":"2026-07-30T05:53:00","date_gmt":"2026-07-30T05:53:00","guid":{"rendered":"https:\/\/pdflove.co\/blog\/?p=163"},"modified":"2026-08-30T02:59:34","modified_gmt":"2026-08-30T02:59:34","slug":"how-to-extract-text-from-an-image","status":"publish","type":"post","link":"https:\/\/pdflove.co\/blog\/how-to-extract-text-from-an-image\/","title":{"rendered":"How to Extract Text from Image Files for Free"},"content":{"rendered":"<p>You have a photo of a page, a screenshot, or a scanned receipt, and you need the words out of it as text you can actually use. To <strong>extract text from image<\/strong> files you need OCR, and with PDFLove you can run it free, in your browser, with no upload and no sign-up.<\/p>\n<p>This guide explains what OCR really does, how to extract text from image files step by step, why results vary so much between one picture and the next, and what to check before you trust the output.<\/p>\n\n<h2>Table of Contents<\/h2>\n<ul>\n  <li><a href=\"#what-ocr-is\">What OCR actually does<\/a><\/li>\n  <li><a href=\"#when\">When you need it<\/a><\/li>\n  <li><a href=\"#how-to\">How to extract text from image files, step by step<\/a><\/li>\n  <li><a href=\"#quality\">Why the source image decides the result<\/a><\/li>\n  <li><a href=\"#vs-retyping\">OCR or just retype it?<\/a><\/li>\n  <li><a href=\"#formats\">Which image formats work<\/a><\/li>\n  <li><a href=\"#phone\">Doing it on a phone<\/a><\/li>\n  <li><a href=\"#desktop\">Doing it on a desktop<\/a><\/li>\n  <li><a href=\"#accuracy\">Checking the output before you trust it<\/a><\/li>\n  <li><a href=\"#limits\">What OCR still gets wrong<\/a><\/li>\n  <li><a href=\"#improve\">Rescuing a difficult image<\/a><\/li>\n  <li><a href=\"#privacy\">Why the upload question matters here<\/a><\/li>\n  <li><a href=\"#problems\">Common problems and how to fix them<\/a><\/li>\n  <li><a href=\"#use-cases\">Real-world use cases<\/a><\/li>\n  <li><a href=\"#why-pdflove\">Why PDFLove is the free, private choice<\/a><\/li>\n  <li><a href=\"#faq\">Frequently asked questions<\/a><\/li>\n<\/ul>\n\n<h2 id=\"what-ocr-is\">What OCR Actually Does<\/h2>\n<p>Before you can extract text from image files, it helps to know that a photograph of a page contains no words. It contains pixels that happen to be arranged in shapes your eye reads as words. Your computer sees a grid of colours and nothing more, which is why you cannot select, search or copy from it.<\/p>\n<p>Optical character recognition is what lets you extract text from image data at all. It finds the regions of the picture that look like text, separates them into lines and characters, matches each shape against learned letterforms, and outputs actual characters. That is what lets you extract text from image files rather than retyping them.<\/p>\n<p>Modern tools that extract text from image files are very good and still not magic. It is reading shapes and making a best guess, so it is confident about clean printed type and much less so about handwriting, unusual fonts, or a photograph taken in poor light. You can read about the history and the underlying approach on <a href=\"https:\/\/en.wikipedia.org\/wiki\/Optical_character_recognition\" rel=\"noopener\" target=\"_blank\">Wikipedia&#8217;s optical character recognition page<\/a>.<\/p>\n\n<h2 id=\"when\">When You Need It<\/h2>\n<p>The situations where you need to extract text from image files repeat more than you would think.<\/p>\n<ul>\n  <li><strong>A screenshot with an error message<\/strong> you want to search for rather than retype into a search box.<\/li>\n  <li><strong>A photographed page from a book or a report<\/strong> that you need to quote.<\/li>\n  <li><strong>A scanned invoice or receipt<\/strong> whose figures need to reach a spreadsheet.<\/li>\n  <li><strong>A PDF that is really a scan<\/strong>, where nothing is selectable and search finds nothing.<\/li>\n  <li><strong>A form someone sent as a picture<\/strong> when you needed the content.<\/li>\n  <li><strong>Old documents being archived<\/strong>, where the point of digitising is being able to search later.<\/li>\n<\/ul>\n<p>In every one of these, the alternative to extract text from image files is retyping, which is slow and introduces its own mistakes.<\/p>\n\n<h2 id=\"how-to\">How to Extract Text from Image Files, Step by Step<\/h2>\n<p>The browser route works the same on Windows, macOS, Linux and ChromeOS, and installs nothing. Here is how to extract text from image files end to end.<\/p>\n<ol>\n  <li><strong>Start with the best copy you have.<\/strong> An original photo beats a screenshot of it, and a screenshot beats a photo of a screen.<\/li>\n  <li><strong>Open the tool.<\/strong> Go to the <a href=\"https:\/\/pdflove.co\/tools\/ocr-pdf\/\">free PDFLove OCR tool<\/a>. Nothing to install, nothing to register.<\/li>\n  <li><strong>Add the file.<\/strong> Drag the image or scanned PDF in. Your browser reads it directly, so it never leaves the device.<\/li>\n  <li><strong>Let it run.<\/strong> Recognition happens locally using your machine&#8217;s own processing power. A dense page takes a few seconds.<\/li>\n  <li><strong>Review the output.<\/strong> Read it against the picture, paying attention to numbers and proper nouns.<\/li>\n  <li><strong>Copy it out, or keep the searchable PDF.<\/strong> For a scan you want to archive, the searchable PDF is usually more useful than raw text, because it keeps the page as it looked.<\/li>\n<\/ol>\n<p>That is all there is to extract text from image files. No account, no watermark on the result, and no page limit.<\/p>\n\n\n<figure class=\"wp-block-image size-large\"><img decoding=\"async\" src=\"https:\/\/pdflove.co\/blog\/wp-content\/uploads\/2026\/08\/how-to-extract-text-from-an-image-example.jpg\" alt=\"Recognised words highlighted as OCR runs to extract text from image files in a browser\"\/><\/figure>\n\n\n<h2 id=\"quality\">Why the Source Image Decides the Result<\/h2>\n<p>Your ability to extract text from image files is decided here, more than anywhere else, and it is almost entirely out of the software&#8217;s hands. OCR can only work with the pixels it is given.<\/p>\n<h3>Resolution<\/h3>\n<p>To extract text from image data, the characters need enough pixels to have a recognisable shape. Around 300 DPI is the usual guidance for a scan; for a photo, the practical test is whether you can read the smallest characters when you zoom in on screen. If you cannot, no OCR engine will.<\/p>\n<h3>Contrast<\/h3>\n<p>Black on white is the ideal when you extract text from image files. Grey on cream, or text over a photograph, gives the engine much less to separate. This is why a scanner&#8217;s black-and-white filter often improves recognition dramatically over the same page in colour.<\/p>\n<h3>Geometry<\/h3>\n<p>Lines should be horizontal. A page shot at an angle produces text on a slant and characters distorted by perspective, both of which hurt. Crop away everything that is not the page.<\/p>\n<h3>Focus and noise<\/h3>\n<p>Blur is fatal in a way nothing else is: a soft photo cannot be sharpened back into readable letterforms. Compression artefacts from a heavily re-saved JPEG have a similar effect on fine detail.<\/p>\n<p>The short version: if you want to extract text from image files reliably, spend the effort at capture rather than hoping the software rescues a bad picture.<\/p>\n\n<h2 id=\"vs-retyping\">Extract Text from Image Files, or Just Retype It?<\/h2>\n<p>An honest question, and the answer is not always to extract text from image files. For four lines on a receipt, retyping is faster than opening anything.<\/p>\n<p>The calculation turns on three things. <strong>Volume<\/strong>: a page is roughly 500 words, and nobody types that faster than a machine reads it. <strong>Accuracy<\/strong>: typing introduces its own errors, and unlike OCR errors they are not concentrated in predictable places you can check. <strong>Repetition<\/strong>: if there are forty receipts, the setup cost of OCR is paid back on the second one.<\/p>\n<p>Where retyping genuinely wins is short, critical, badly captured text &#8211; a handwritten note, a serial number on a curved label, four figures from a crumpled ticket. In those cases you would be verifying every character of the OCR output anyway, so the machine has saved you nothing.<\/p>\n<p>The rule of thumb: if it is more than a few lines, or there is more than one of it, extract text from image files rather than typing. Below that, type it.<\/p>\n\n<h2 id=\"formats\">Which Image Formats Work<\/h2>\n<p>Attempts to extract text from image files do not much care about the container, but it cares a great deal about what the container did to the pixels.<\/p>\n<ul>\n  <li><strong>PNG<\/strong> is the best case for screenshots. It is lossless, so letterforms keep their edges exactly as rendered.<\/li>\n  <li><strong>JPEG<\/strong> is fine at high quality and poor at low. Its compression blurs precisely the fine detail OCR depends on, and every re-save compounds it.<\/li>\n  <li><strong>HEIC<\/strong> from an iPhone works, though converting to PNG or a high-quality JPEG first avoids any support gaps.<\/li>\n  <li><strong>WebP and AVIF<\/strong> are increasingly common and behave like JPEG &#8211; good at high quality, harmful at low.<\/li>\n  <li><strong>PDF<\/strong> may already contain real text. Try selecting a word first; if you can, you do not need OCR at all.<\/li>\n<\/ul>\n<p>The practical warning is about provenance rather than format. A picture that has been screenshotted, sent through a chat app, saved, and forwarded again has been recompressed at each step. By the time you try to extract text from image data that has made that journey, much of the detail the engine needs is simply gone. Always go back to the original if you can find it.<\/p>\n\n<h2 id=\"phone\">Doing It on a Phone<\/h2>\n<p>Phones are well suited to extract text from image captures, because the camera and the recognition sit on the same device.<\/p>\n<p>The <a href=\"https:\/\/pdflove.co\/mobile-app\/\">free PDFLove app<\/a> scans and runs OCR on the device in more than twenty languages, so you can photograph a page and get searchable text without anything being transmitted.<\/p>\n<p>Both iOS and Android also have built-in text selection in the camera or photo viewer, which is genuinely handy for grabbing a phone number or a line of an address. It is less suited to a full page, where you want the result as a document rather than a selection you have to paste somewhere.<\/p>\n\n<h2 id=\"desktop\">Doing It on a Desktop<\/h2>\n<p>To extract text from image files on a computer the file is usually already saved, so the browser tool is the shortest path: open it, drop the file, read the result.<\/p>\n<p>Desktop OCR software exists and is capable, particularly for very large batches, but it is a heavier commitment &#8211; an install, often a licence, and for scanned archives a learning curve. For the ordinary case of one image or one scanned PDF, it is far more machinery than the job needs.<\/p>\n<p>The other desktop route worth knowing is that if your PDF is a scan, running OCR over it in place is usually better than pulling the words out. You keep the page exactly as it looks and gain the ability to search it.<\/p>\n\n<h2 id=\"accuracy\">Checking the Output Before You Trust It<\/h2>\n<p>Output looks authoritative when you extract text from image files, because it arrives as neatly formatted text. It is still a guess, and the failures are quiet.<\/p>\n<ul>\n  <li><strong>Check every number.<\/strong> Digits carry no context to help the engine, so this is where errors do real damage &#8211; an account number or a total is wrong silently.<\/li>\n  <li><strong>Watch the classic confusions.<\/strong> 0 and O, 1 and l and I, 5 and S, 8 and B, rn read as m.<\/li>\n  <li><strong>Check proper nouns.<\/strong> Names and addresses cannot be corrected from a dictionary, so they are guessed less reliably than ordinary words.<\/li>\n  <li><strong>Look at the layout.<\/strong> Columns, tables and footnotes often come out in the wrong reading order even when every character is right.<\/li>\n  <li><strong>Read the last lines.<\/strong> The bottom of a page is the most likely to be cropped, shadowed or out of focus.<\/li>\n<\/ul>\n<p>For anything financial or legal, treat what you extract text from image files as a draft to verify rather than a result to paste.<\/p>\n\n<h2 id=\"limits\">What OCR Still Gets Wrong<\/h2>\n<p>Being clear about the boundaries of any attempt to extract text from image files saves a lot of frustration.<\/p>\n<p><strong>Handwriting<\/strong> is a different problem from printed text, and general OCR handles it poorly. Neat block capitals sometimes work; ordinary cursive usually does not.<\/p>\n<p><strong>Tables<\/strong> survive as characters but rarely as structure. Expect to rebuild the columns yourself.<\/p>\n<p><strong>Multi-column layouts<\/strong> can be read straight across the page rather than down each column, producing text that is accurate word by word and nonsense as a whole.<\/p>\n<p><strong>Decorative and very condensed fonts<\/strong>, and any text set over a busy background, degrade sharply.<\/p>\n<p><strong>Mixed languages on one page<\/strong> confuse an engine set to a single language, particularly across different scripts.<\/p>\n<p>None of these are reasons to avoid OCR. They are reasons to look at what came out.<\/p>\n\n<h2 id=\"improve\">How to Extract Text from Image Files the Software Struggles With<\/h2>\n<p>Sometimes the picture is all you have and it is not good. A few interventions genuinely help, and they are all about giving the engine cleaner pixels rather than asking more of it.<\/p>\n<ol>\n  <li><strong>Crop to the text.<\/strong> Remove borders, desk, fingers and anything else. A tighter frame means fewer regions to misidentify as characters.<\/li>\n  <li><strong>Do one column at a time.<\/strong> Multi-column pages are the commonest cause of accurate words in nonsensical order. Crop each column and run it separately.<\/li>\n  <li><strong>Straighten it.<\/strong> Even a few degrees of rotation costs accuracy, because the engine expects horizontal baselines.<\/li>\n  <li><strong>Raise the contrast.<\/strong> Pushing grey-on-cream towards black-on-white is the single most effective adjustment, which is why a scanner&#8217;s black-and-white filter helps so much.<\/li>\n  <li><strong>Scale up a small capture<\/strong> before running it. Doubling the size of a low-resolution crop sometimes gives the engine enough shape to work with, though it cannot invent detail that was never captured.<\/li>\n  <li><strong>Set the right language<\/strong> before you start, especially for accented or non-Latin text.<\/li>\n  <li><strong>Recapture if you can.<\/strong> Thirty seconds with better light beats an hour of fighting a bad photograph, every time.<\/li>\n<\/ol>\n<p>What none of this can do is recover focus. Blur is the one failure with no remedy, because the information was never recorded. If you cannot read it, the software cannot either, and no amount of processing changes that.<\/p>\n\n<h2 id=\"privacy\">Why the Upload Question Matters Here<\/h2>\n<p>Consider what people actually extract text from image files for: invoices, payslips, medical letters, contracts, identity documents, screenshots of private conversations. It is a fair sample of everything sensitive you own.<\/p>\n<p>Most online OCR services upload the file to a server, process it there and send the text back. Your document has then been copied onto infrastructure you cannot inspect, governed by a retention policy you did not read.<\/p>\n<p>PDFLove runs recognition inside your browser using your own device. Nothing is transmitted, so there is no server-side copy to leak or retain. For this particular task that is not a bonus feature &#8211; it is the reason to choose the tool.<\/p>\n\n<h2 id=\"problems\">Common Problems and How to Fix Them<\/h2>\n<p>These are the snags that come up most often, and most trace back to the picture rather than the software.<\/p>\n<h3>The output is gibberish<\/h3>\n<p>Almost always the source. Blur, low resolution or poor contrast. Recapture the image rather than trying to clean up the text.<\/p>\n<h3>It missed whole sections<\/h3>\n<p>Those regions were probably not recognised as text at all &#8211; white on a dark background, text over a photo, or a heavily stylised font. Crop to the section and try it alone.<\/p>\n<h3>The words are right but the order is wrong<\/h3>\n<p>A multi-column layout read across instead of down. Crop each column separately and run them one at a time.<\/p>\n<h3>Numbers are wrong<\/h3>\n<p>The commonest and most damaging failure. Always verify digits by eye; a higher-resolution capture usually helps.<\/p>\n<h3>My PDF still is not searchable<\/h3>\n<p>It is a scan with no text layer. Run OCR over the PDF itself rather than extracting the words, and you keep the appearance and gain the search.<\/p>\n<h3>Accented or non-Latin characters come out wrong<\/h3>\n<p>The engine is set to the wrong language. Choose the right one; PDFLove supports more than twenty.<\/p>\n\n<h2 id=\"use-cases\">Real-World Use Cases<\/h2>\n<ul>\n  <li><strong>Getting receipt totals into a spreadsheet<\/strong>, the commonest reason to extract text from image files at all, without retyping each one.<\/li>\n  <li><strong>Quoting from a photographed page<\/strong> of a book, report or article.<\/li>\n  <li><strong>Searching an archive of old scans<\/strong> that has never been searchable.<\/li>\n  <li><strong>Pulling an error message out of a screenshot<\/strong> so you can search for it.<\/li>\n  <li><strong>Making documents accessible<\/strong>, since a screen reader gets nothing from an image of text.<\/li>\n  <li><strong>Recovering content<\/strong> from a form or letter that only exists as a picture.<\/li>\n<\/ul>\n\n<h2 id=\"why-pdflove\">Why PDFLove Is the Free, Private Choice<\/h2>\n<p>Free tools to extract text from image files usually come with conditions: a page limit, a daily cap, a watermark, an account, or an upload you were not really told about. The service has costs, and those are the ways they are recovered.<\/p>\n<p>PDFLove lets you extract text from image files with none of them. The tool is free outright, there is no account or email capture, no watermark and no page cap. Because recognition runs on your device rather than on a server, there is no per-page cost to pass on in the first place.<\/p>\n<p>That architecture is also what makes the privacy claim real rather than a promise. Your file is opened by your browser and processed there; it is never transmitted. When you extract text from image files this way, the picture and the words both stay on your machine. The rest of the <a href=\"https:\/\/pdflove.co\/tools\/\">free PDF toolkit<\/a> works identically.<\/p>\n\n<h2 id=\"faq\">Frequently Asked Questions<\/h2>\n<div id=\"rank-math-faq\" class=\"rank-math-block\">\n<div class=\"rank-math-list \">\n<div id=\"faq-question-1\" class=\"rank-math-list-item\">\n<h3 class=\"rank-math-question \">How do I extract text from image files for free?<\/h3>\n<div class=\"rank-math-answer \">\n\n<p>Open the <a href=\"https:\/\/pdflove.co\/tools\/ocr-pdf\/\">free PDFLove OCR tool<\/a>, drag in your image or scanned PDF, and let it run. Recognition happens in your browser on your own device, so nothing is uploaded. There is no account, no watermark and no page limit.<\/p>\n\n<\/div>\n<\/div>\n<div id=\"faq-question-2\" class=\"rank-math-list-item\">\n<h3 class=\"rank-math-question \">Why is the extracted text full of mistakes?<\/h3>\n<div class=\"rank-math-answer \">\n\n<p>Nearly always the source image. OCR can only work with the pixels it is given, so blur, low resolution, poor contrast or a steep angle all degrade it badly. If you cannot read the smallest characters by zooming in on screen, no OCR engine will either. Recapture rather than trying to fix the text.<\/p>\n\n<\/div>\n<\/div>\n<div id=\"faq-question-3\" class=\"rank-math-list-item\">\n<h3 class=\"rank-math-question \">Can OCR read handwriting?<\/h3>\n<div class=\"rank-math-answer \">\n\n<p>Generally no. Handwriting recognition is a different problem from printed text, and general-purpose OCR handles it poorly. Neat block capitals sometimes come through; ordinary cursive usually does not.<\/p>\n\n<\/div>\n<\/div>\n<div id=\"faq-question-4\" class=\"rank-math-list-item\">\n<h3 class=\"rank-math-question \">Is it safe to use an online OCR tool?<\/h3>\n<div class=\"rank-math-answer \">\n\n<p>It depends on the tool. Most upload your file to a server to process it, which means invoices, payslips and identity documents are copied onto infrastructure you cannot inspect. PDFLove runs recognition inside your browser, so the file never leaves your device.<\/p>\n\n<\/div>\n<\/div>\n<div id=\"faq-question-5\" class=\"rank-math-list-item\">\n<h3 class=\"rank-math-question \">What resolution do I need?<\/h3>\n<div class=\"rank-math-answer \">\n\n<p>Around 300 DPI for a scan. For a photograph, the practical test is whether the smallest text is legible when you zoom in on screen. Bigger is not always better &#8211; past a point you are adding file size rather than detail.<\/p>\n\n<\/div>\n<\/div>\n<div id=\"faq-question-6\" class=\"rank-math-list-item\">\n<h3 class=\"rank-math-question \">Should I extract the text or make the PDF searchable?<\/h3>\n<div class=\"rank-math-answer \">\n\n<p>For a scanned document you want to keep, adding a text layer is usually better: the page still looks exactly as it did, and it becomes searchable and readable by screen readers. Extract raw text when you need the words somewhere else, such as in a spreadsheet or an email.<\/p>\n\n<\/div>\n<\/div>\n<div id=\"faq-question-7\" class=\"rank-math-list-item\">\n<h3 class=\"rank-math-question \">Why did my table come out scrambled?<\/h3>\n<div class=\"rank-math-answer \">\n\n<p>OCR recognises characters, not structure. Tables usually survive as text but lose their columns, and multi-column pages are sometimes read straight across instead of down. Crop each column and run it separately for a cleaner result.<\/p>\n\n<\/div>\n<\/div>\n<div id=\"faq-question-8\" class=\"rank-math-list-item\">\n<h3 class=\"rank-math-question \">Does it work in languages other than English?<\/h3>\n<div class=\"rank-math-answer \">\n\n<p>Yes. PDFLove supports more than twenty languages. Choose the right one before running it, because an engine set to the wrong language mangles accented and non-Latin characters.<\/p>\n\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n\n<h2>Conclusion<\/h2>\n<p>Being able to extract text from image files removes one of the more tedious jobs in digital paperwork. Words locked inside a picture become words you can search, copy, correct and quote.<\/p>\n<p>To extract text from image files well: start with the sharpest, squarest, highest-contrast source you can get, run OCR, then read the output against the original with particular attention to numbers and names. That last step is what separates useful text from confidently wrong text.<\/p>\n<p>Open the <a href=\"https:\/\/pdflove.co\/tools\/ocr-pdf\/\">free OCR tool<\/a> and get the words out. Free, private, and never uploaded anywhere.<\/p>","protected":false},"excerpt":{"rendered":"<p>You have a photo of a page, a screenshot, or a scanned receipt, and you need the words out of it as text you can&#8230;<\/p>\n","protected":false},"author":2,"featured_media":161,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"rank_math_focus_keyword":"extract text from image","rank_math_title":"Extract Text from Image Free: 4 Easy Fast Ways 2026","rank_math_description":"Learn how to extract text from image files for free with OCR. Copy, search and edit words from a photo or scan - private, in your browser, no upload.","_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[7],"tags":[],"class_list":["post-163","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-scan-ocr"],"jetpack_featured_media_url":"https:\/\/pdflove.co\/blog\/wp-content\/uploads\/2026\/08\/how-to-extract-text-from-an-image.jpg","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/pdflove.co\/blog\/wp-json\/wp\/v2\/posts\/163","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/pdflove.co\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/pdflove.co\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/pdflove.co\/blog\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/pdflove.co\/blog\/wp-json\/wp\/v2\/comments?post=163"}],"version-history":[{"count":1,"href":"https:\/\/pdflove.co\/blog\/wp-json\/wp\/v2\/posts\/163\/revisions"}],"predecessor-version":[{"id":172,"href":"https:\/\/pdflove.co\/blog\/wp-json\/wp\/v2\/posts\/163\/revisions\/172"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/pdflove.co\/blog\/wp-json\/wp\/v2\/media\/161"}],"wp:attachment":[{"href":"https:\/\/pdflove.co\/blog\/wp-json\/wp\/v2\/media?parent=163"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/pdflove.co\/blog\/wp-json\/wp\/v2\/categories?post=163"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/pdflove.co\/blog\/wp-json\/wp\/v2\/tags?post=163"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}