Consider using SubtlexCN data

[hugolpz]

• SUBTLEX-CH: Chinese Word and Character Frequencies Based on Film Subtitles

Arguably better quality for teaching than Junda written corpora.

See also #52

[benlieb]

@hugolpz I love this idea. I'm not sure this repo is under active maintenance atm, though. I'd be open to contributing and maybe helping with this integration, or putting together another npm package.

@hugolpz any interest?