words_and_pictures

Python and Javascript to make and play talking pictures.

Create simple, interactive reading lessons with words and pictures by creating three things:

• pictures
• words
• audio files of/from the words

The pictures are grouped on slides.

The words (or phonemes) are separated into blocks.

See demo with 16 slides, telling a "Pink Shoes" adventure story.

Production time on these stories is remarkably short. Pink Shoes 2 was started at 4 pm and completed by 7:15 pm.

Other demos show presentation of phonemes and multiple languages.

##JSON Format for Linking Pictures (img) to Words (text)

Example input (test.json):

{
  "block-type" : "words",
  "language-default" : "en-uk",
  "slides" : [
    {
      "slide" : [
        {
          "img" : "img1.png",
          "text" : "This is a"
        },
        {
          "img" : "img2.png",
          "text" : "demo slide."
        }
      ]
    }
  ],
  "tts-engine" : "google"
}

Use

$ python json2mp3.py test.json

to use the Google Translate Text-to-Speech API to create .mp3 files into the audio folder (subfolder <your json file name>).

On a Mac, use "tts-engine" : "say" to use the say utility and ffmpeg to create .mp3 files from the words and add the file names (with key of "audio") to each block. Output goes into test_output.json or (<your file name stem>_output.json).

Prefer your own voice? Use a sound file editor to create .mp3 files with names following the pattern of s + slide_block + .mp3:

• s0_0.mp3 for full audio of first slide
• s0_1.mp3 for audio of first block of words on first slide
• s0_2.mp3 for audio of second block of words on first slide
• ...
• s1_0.mp3 for full audio of second slide
• s1_1.mp3 for audio of first block of words on second slide
• s1_2.mp3 for audio of second block of words on second slide

##Playing with Cloned Repo

With Python 2:

$ git clone https://github.com/jg1141/jg1141.github.io.git
$ cd jg1141.github.io/
$ python -m SimpleHTTPServer

Open your web browser to http://localhost:8000/.