Topic guide · Cluster B
Convertir un PDF vers et depuis d'autres formats — sans envoi de fichier
Convertir des PDF en JPG, PNG, texte brut et Markdown, transformer des images en PDF, exécuter l'OCR sur des pages scannées, discuter avec un document grâce à l'IA, compresser pour l'envoi par e-mail, et comparer des versions. Tout dans le navigateur.
La plupart des documents n'arrivent pas dans le format dont vous avez besoin. Un contrat scanné est un empilement d'images qu'il faut pouvoir lire comme du texte. Un catalogue produit arrive en PDF et l'équipe design a besoin d'un JPG de chaque page pour la diffusion sur les réseaux. Un long mémoire juridique est plus simple à résumer quand on peut poser la question en français courant. Un rapport est devenu trop lourd pour être envoyé par e-mail et il vous faut une copie plus légère qui reste lisible. La boîte à outils Convertir est le pont entre le PDF et les formats qui l'entourent. Cette page rassemble tous les outils PdfWiseAI qui changent le format d'un document plutôt que sa structure interne : transformer des pages PDF en images, transformer des images en PDF, extraire le texte lisible d'un PDF, exécuter l'OCR sur des pages composées d'images, poser une question à un document via l'IA, réduire la taille d'un PDF pour l'envoi, et comparer deux versions d'un document côte à côte. L'export d'images, l'export de texte et l'OCR s'exécutent entièrement dans le navigateur. Le chat IA est la seule opération qui envoie du texte à un service distant, et uniquement le texte nécessaire à la réponse — le PDF binaire n'est pas transmis. Si vous savez dans quel format vous voulez aller, sautez directement à la section correspondante. Si vous ne savez pas si un fichier est un vrai PDF ou la numérisation d'un document, la section OCR explique comment faire la différence et que faire ensuite.
PDF to Images
How to convert PDF pages to JPG or PNG
Export every page of a PDF as a high-quality image. Useful for presentations, design previews, social posts, thumbnails and any system that accepts images but not PDF documents. Choose JPG for small file size or PNG when you need pixel-exact rendering with no compression artefacts, then pick a quality level appropriate to the destination. Rendering runs locally, so the source PDF does not pass through a third-party conversion server.
Try PDF to Images →Images to PDF
How to convert JPG and PNG images to PDF
Combine photos, scans, receipts or screenshots into a single PDF. Each image becomes one page, in the order you select, and the result is a PDF that is easier to share, archive and annotate than a folder of loose files. Useful for expense reports, scan bundles, portfolios and application documents. The PDF is created in the browser; the source images are not uploaded.
Try Images to PDF →PDF to Text
How to extract text from a PDF
Pull the readable content of a PDF out as a plain text file, with page boundaries preserved so the output can be traced back to the source. Useful for notes, analysis, search and reuse. The tool works best with digitally generated PDFs that contain selectable text. Scanned, image-only documents need OCR first — see the OCR section below for that workflow.
Try PDF to Text →OCR PDF
How to OCR a scanned PDF
Run optical character recognition on image-only PDFs in the browser. Tesseract.js renders each page at 2× resolution and detects the text, producing a page-by-page text report you can download. Useful for digitising paper records, making scanned documents searchable, and unlocking text that an image viewer cannot select. If a PDF already has selectable text, OCR is skipped automatically to save time.
Try OCR PDF →Chat with PDF
How to chat with a PDF using AI
The only operation on this page that does not run entirely in the browser. PdfWiseAI reads the text out of the PDF locally, then sends the relevant text plus your question to an AI service. The binary PDF never leaves your device. Useful for summarising a long contract, extracting amounts and dates, comparing two clauses, or producing a structured table from a long document. Always verify the cited pages before relying on the answer.
Try Chat with PDF →Compress PDF
How to compress a PDF for sharing
Reduce the file size of a PDF before emailing it, sharing it through a chat app or uploading it to a portal. The compression tool offers several levels so you can balance file size against image quality — start with the recommended setting and step up to stronger compression only if needed. Runs locally, so the source PDF is not uploaded. Best results come from documents where images dominate the size; text-only PDFs are usually already small.
Try Compress PDF →Compare PDFs
How to compare two PDF versions
When you have two versions of a document and need to know what changed, the compare tool renders each page of both PDFs and computes a pixel-level diff. The on-screen summary lists every page that differs and the percentage of change, and the downloadable report highlights the differences in red. Useful for reviewing redlines, spotting a bad save, or auditing changes between revisions.
Try Compare PDFs →Frequently asked questions
- 01Can I convert a PDF to JPG without uploading it?
- Yes. The PDF to images tool renders each page in the browser and packages the results in a ZIP. The source PDF is not sent to a conversion server. You can pick JPG for small file size or PNG when you need lossless rendering, and choose a quality level appropriate to your use.
- 02Do I lose quality when converting PDF to image?
- You can choose. The standard quality is a good balance for most uses, including social posts and previews. For print or design work, pick the high or very high option, which produces a larger file with no visible artefacts. The original PDF is not changed — you can re-export at a different quality any time.
- 03What is the difference between PDF to text and OCR?
- PDF to text reads the selectable text layer that exists in digitally generated PDFs. OCR runs on image-only pages (scans) and recognises the text from the pixels. If you are not sure which one to use, open the PDF in any viewer and try to select a word with the mouse — if you can highlight it, use PDF to text; if you cannot, run OCR.
- 04Can I extract text from a scanned PDF?
- Yes, by running OCR first. OCR adds a text layer to the scanned pages, which PDF to text (and any other tool that relies on text selection, including search) can then read. The OCR tool runs entirely in the browser with Tesseract.js.
- 05How do I reduce a PDF file size without losing too much quality?
- Start with the recommended compression level. If the result is still too large, try the stronger level — image quality will drop a little, but text stays sharp. For scanned documents, the biggest win usually comes from re-OCRing at a lower DPI rather than re-compressing.
- 06Is the chat tool different from the other tools on this page?
- Yes. The other six operations on this page run entirely in your browser. The chat tool reads the text locally and sends the necessary text plus your question to an AI service. The binary PDF is not transmitted, but the extracted text of the relevant pages is. Review the AI provider's data policy if that matters for your document.