podcast_tools

A python-based tool suite for podcast. Tools include:

1. creating audiogram:
   • creates an audiogram from audio input for YouTube
   • runs in parallel using multiprocessing
   • handles Chinese/English mixed text justification (see example below)

Dependencies

python libraries:

• matplotlib
• scipy
• numpy
• moveipy
• pydub
• mutagen

Additional:

• ffmpeg

Examples

creating audiograms

As shown in the examples/audiogram directory, given the following

• input.mp3: the input audio file
• podcast_logo.png: the podcast logo
• specs.json: a general podcast spec defining colors, box sizes, etc
• ep_specs.json: a more specific episode spec defining episode title and font size. The episode can be chinese/english mixed text.

the module /audiogram/audiogram.py provides the capability to create an audiogram, as shown below. Notably, the /audiogram/text_util.py module will fit Chinese/English mixed text into the box with the correct line width.

output_audiogram.mp4