search

DrewThomasson / ebook2audiobook

Generates an audiobook with chapters and ebook metadata using Calibre and Xtts from Coqui tts, and with optional voice cloning, and supports multiple languages
MIT License
795 stars 78 forks source link

The output m4b doesn't work #21

Closed nitinmukesh closed 1 month ago

nitinmukesh commented 1 month ago

So, I run the program using docker and used webUI to convert small 2 page PDF

image

It showed conversion is complete. I downloaded the file using the download icon at the top-right of Audiobook Player.

Now I tried playing using VLC, also tried converting using online convertors. Nothing is working.

How to use this file. (attached the output as zip) comic.zip

Log

Output #0, ipod, to './Audiobooks/comic.m4b': Metadata: encoder : Lavf59.27.100 Stream #0:0: Audio: aac (LC) (mp4a / 0x6134706D), 7350 Hz, mono, fltp, 44 kb/s Metadata: encoder : Lavc59.37.100 aac Stream #0:1: Video: png, rgb24(pc, gbr/unknown/unknown, progressive), 1275x1650 [SAR 1:1 DAR 17:22], q=2-31, 200 kb/s, 25 fps, 12800 tbn (attached pic) Metadata: encoder : Lavc59.37.100 png frame= 1 fps=0.0 q=-0.0 Lsize= 1517kB time=00:00:00.00 bitrate=159274051.3kbits/s speed=0.000509x video:1516kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.018163% [aac @ 0x5592a7cc5880] Qavg: nan Audiobook created at ./Audiobooks/comic.m4b
DrewThomasson commented 1 month ago

It's probs because your input pdf has no text in it, considering it's a pdf of a comic book.

Your comic pdf is most likely a set of images of pages of the comic book.

So when it uses Calibre to convert your comic pdf file, you get an output txt file that contains no text.

nitinmukesh commented 1 month ago

Oh ok. Got it. I thought it does OCR.

DrewThomasson commented 1 month ago

Here's a huggingface space that'll to the PDF OCR to TXT step for you tho lol.

https://huggingface.co/spaces/drewThomasson/PDF-to-TXT-OCR?logs=build