Photo to speech: read the text in a picture aloud

Step one: get the text out of the photo

Check the recognized text once: OCR mistakes such as broken words or swapped letters are read aloud exactly as they are.

• On an iPhone, Live Text lets you select and copy text in photos and in the camera view.
• On Android, Google Lens selects and copies text from a photo or the camera.
• On a computer, an OCR tool such as Tesseract turns an image into a text file.
• Good light, a straight angle and sharp focus matter more than the app you use.

Step two: voice it free online

Paste a sentence of 30 to 100 characters into the page of any voice, choose the language and press Voice it. In a few seconds you can listen and download the recording, with no account.

Clean up the text before voicing

Copy one paragraph at a time from the photo: it is easier to check than a whole page.

• Join lines that a page broke in the middle of a sentence; a line break adds no pause anyway.
• Remove page numbers, headers and footnote marks.
• Fix hyphenated words split across lines.
• Write numbers and abbreviations the way they should be read.

Whole pages and many photos

For a whole page, recognize it with Tesseract, for example tesseract photo.jpg page -l eng, which writes page.txt. Then install the free voice engine with python -m pip install superclonica-cpu and run superclonica --text-file page.txt --voice F2 --lang en --normalize --output page.mp3. The engine has no character limit and runs offline, so a scanned chapter or a stack of worksheets becomes audio without uploading anything. Name the files by page number to keep them in order.

Privacy and what you may voice

Both steps can run on your own devices, so private letters and documents stay private; only the free online test sends the sentence you type to SuperClonica. Voicing a book page for your own listening is personal use, while publishing the audio of a book needs the rights holder’s permission.