ARPABET Encoding

[prlabu]

Awesome resource!

I think a useful annotation to add might be ARPABET encodings for each phoneme. I suspect a lot of people who are interested in a machine-readable format would also be using the CMU Dictionary and other ARPABET-encoded speech recognition software.

So

    "t": {
        "name": "voiceless alveolar stop",
        "arpabet": "T"
        "features": {...}
      }