Books
Your book has no audiobook: how to choose a listening option
If your book has no audiobook, you can check for recordings of the specific edition or use speech synthesis with an electronic text you have the right to use. The choice depends on whether you want a particular performance or simply to hear the contents. TGRadio reads supplied text aloud; it does not guarantee that a particular title is in its catalogue.
Identify the recording you actually need
Check the author, full title, language, and translator. If a particular edition matters, include its publisher or edition number. A matching title does not guarantee matching content: a recording may use another translation, an abridged text, or only part of the work.
Check information from the publisher or author. As a hypothetical example, you might need the complete book in a particular translation while the recording you find contains selected chapters. That recording would not replace the required text. If an actor's interpretation is essential, speech synthesis will not meet that need either.
Check whether you have a usable source
Speech synthesis needs accessible electronic text. Use a book you have the right to use. A cover image, a description, or a purchase page does not supply the book's text. If you do not yet have a file, first obtain a suitable electronic edition.
TGRadio accepts EPUB, FB2, TXT, and PDFs with a text layer. Files can be up to 10 MB, with up to 2 million characters of extracted text per book. A paper copy or an image-only scan needs separate text preparation outside TGRadio, which has no built-in OCR.
Check completeness and readability
Open the beginning, middle, and end of the file before importing it. Compare the contents list with the text and check chapter endings. For a linked book, confirm that the page contains all the text you need. Importing one page does not automatically follow subsequent chapters or pagination.
Correct recognition errors, broken words, and repeated page headers in an external editor before uploading. Saving flawed text in another format does not improve it. TGRadio has no built-in text or pronunciation editor, so prepare a source with noticeable errors beforehand.
Decide whether speech synthesis suits the material
Inspect a few representative pages. For continuous prose, consider whether a single reading voice is sufficient. If meaning depends on diagrams, formulas, or the placement of table entries, an audio track may convey only part of the material. TGRadio does not recognise images or formulas.
Do not expect separate voices for characters in dialogue. As a hypothetical example, a play's character names and lines may all be read in the same selected voice. If you want a production with a cast, keep looking for that kind of recording.
Try a fragment before committing to the book
Add the prepared file and listen to the first ready fragment. Compare it with the source, especially names and difficult passages. Text fragments are sent to a speech synthesis service, so take that into account when choosing material. Personal uploads are not automatically made public.
Book narration is currently free, with no monthly character limit. Technical file and library limits still apply. If the sample reveals source errors, correct the text outside the app first. If you simply prefer another voice, changing it requires new audio to be prepared.