TGRadioBooks
No text found in a PDF for audio: what to check
If no text is found in a PDF for audio, first copy an ordinary paragraph into a text editor. The visible letters may belong to an image, or the text layer may be empty or damaged. Opening the document confirms that its pages can be displayed, but does not confirm that usable text can be extracted.
Copy a paragraph from the page
Find a page of the main text and select several lines within a paragraph. Paste them into a plain text editor. Check the words, spaces, and sentence order. Copying the document title from its properties tells you nothing about the text on its pages.
If selection produces a rectangular image area and pasting gives an image or nothing, the page may be a scan. Also try searching for a distinctive word visible on that page. Failed search and copying together suggest inaccessible text, although they do not establish the exact cause.
Test several parts of the book
Repeat the copying test near the beginning, middle, and end. One PDF can combine text pages, scanned appendices, and an image-only cover. A readable contents page does not establish that the main body is extractable.
For example, the first ten pages might copy correctly while the following chapter consists of images. Note where the result changes. TGRadio does not recognize images or formulas, so importing the same file again will not supply their missing text.
Inspect the pasted wording
Text may be selectable yet paste as strange symbols, lose spaces, or appear in the wrong order. Such a PDF contains accessible characters, but its extracted text may still be unsuitable for listening. A well-rendered page does not resolve extraction problems.
Compare the pasted paragraph with the page word by word. For example, inspect a name, a number, and the end of a long line. If errors are widespread, request another text edition. Changing the voice cannot restore letters that were extracted incorrectly.
Rule out an incomplete or protected file
Close the document and reopen the saved file. Inspect its final pages for missing content or a damage warning. If the copy appears incomplete, obtain it again from an authorized source and repeat the paragraph test.
Check separately for passwords and copying restrictions. TGRadio does not support password-protected PDFs; ask the owner for an authorized unprotected version. Files must also fit the 10 MB file limit and the limit of 2 million extracted characters per book.
Prepare readable text outside the application
A scan needs a text edition or OCR in an external tool. Proofread several complete paragraphs and difficult layouts afterward. This is a quality check, not a stated minimum length for import. Recognizing one line does not make the whole book usable.
TGRadio has no built-in OCR or text editor. Upload the prepared file, select the book and voice, then press Play. Check a familiar paragraph by listening. If import still fails, give support the exact error message and the results of your copying tests.