Dot Crepe uses Open JTalk for Japanese speech synthesis. Voice: HTS Voice "Mei" (mei_normal) MMDAgent Project Team, Nagoya Institute of Technology https://github.com/r9y9/pyopenjtalk/tree/master/pyopenjtalk/htsvoice Creative Commons Attribution 3.0: https://creativecommons.org/licenses/by/3.0/ The voice model is used to generate the announcements. The fixed calls are gain-normalized, resampled and encoded as MP3. See licenses/mei.txt. Runtime: @keanu-thakalath/openjtalkjs 0.1.0 https://github.com/keanu-thakalath/openjtalkjs Browser and WebAssembly files are included, with changes to compressed dictionary loading, audio buffer handling, and synthesis time limits. See licenses/openjtalkjs.txt. Open JTalk, HTS Engine, MeCab and their dictionary data https://open-jtalk.sourceforge.net/ See licenses/open-jtalk.txt, licenses/hts-engine.txt, licenses/mecab.txt and licenses/dictionary.txt. WebAssembly runtime: Emscripten https://github.com/emscripten-core/emscripten See licenses/emscripten.txt.