@Voice works with PDFs, too, and supports neural voices such as Amazon Polly at no extra cost. See this comment of mine. However, there are a few annoyances:
The neural voices stop working eventually, for inexplicable reasons. You can re-enable them, but they will stop working again shortly thereafter. When this happens, @Voice switches to the default voice, rather than stopping playback altogether. So this issue doesn’t prevent you from listening to the book/article—but it means you won’t be able to always do so with a nice-sounding voice.
The conversion to audio doesn’t remove all extraneous elements from the text, especially in books, such as headers, footers, (sometimes) page numbers, inline citations, etc. Nor does it handle footnotes properly: books with footnotes—as opposed to endnotes—are just unlistenable.
@Voice works with PDFs, too, and supports neural voices such as Amazon Polly at no extra cost. See this comment of mine. However, there are a few annoyances:
The neural voices stop working eventually, for inexplicable reasons. You can re-enable them, but they will stop working again shortly thereafter. When this happens, @Voice switches to the default voice, rather than stopping playback altogether. So this issue doesn’t prevent you from listening to the book/article—but it means you won’t be able to always do so with a nice-sounding voice.
The conversion to audio doesn’t remove all extraneous elements from the text, especially in books, such as headers, footers, (sometimes) page numbers, inline citations, etc. Nor does it handle footnotes properly: books with footnotes—as opposed to endnotes—are just unlistenable.