pdf2speech: generate audio tracks from text 📄 ➡️ 🔊

Reading PDF files and converting them to audible cues. Feed your favorite books and rest you eyesight while going back home after a hard day of work and reading documents on the computer screen.

Install

git clone https://github.com/Aavache/pdf2speech.git
cd pdf2speech
pip install .

How to use the tool

Ask for help to the terminal and insert the requested arguments:

p2s --help

>>>
Usage: p2s [OPTIONS] TEXT_PATH OUTPUT_DIR

  Convert a text file into speech audio

Options:
  -fp, --from_page INTEGER       From which page to start
  -v, --vocoder [gtts]           Vocooder name  [default: gtts]
  -w, --words-per-track INTEGER  Number of pages per track  [default: 500]
  -l, --language TEXT            Language  [default: en]
  --help                         Show this message and exit.

Take a look to the examples in input_examples and the corresponding generated audios tracks outputs.

Reference

The code is generic to any text to speech vocoder, however at the moment only the google-based synthesize is available, i.e. gTTs.

Future work

• Support state-of-the-art TTS methods.
• Supporting more text formats, at the moment pdf and txt are allowed.