jpreprocess

jpreprocess (python binding)

Japanese text preprocessor for Text-to-Speech application.

This is a python binding of jpreprocess, which is written in Rust. The rust library is published in crates.io.

JPreprocess (the base code written in Rust) is a rewrite of OpenJTalk.

Usage

Unlike pyopenjtalk, this package does not include support of marine and TTS.

Currently, this package is for text processing only.

Run text processing frontend

import jpreprocess

j = jpreprocess.jpreprocess()
njd_features = j.run_frontend("本日は晴天なり")

assert njd_features[0].get("string") == "本日"
assert njd_features[0].get("pos") == "名詞"

Extract full-context label

import jpreprocess

j = jpreprocess.jpreprocess()
fullcontext = j.extract_fullcontext("本日は晴天なり")

assert len(fullcontext) == 21
assert fullcontext[0] == r"xx^xx-sil+h=o/A:xx+xx+xx/B:xx-xx_xx/C:xx_xx+xx/D:xx+xx_xx/E:xx_xx!xx_xx-xx/F:xx_xx#xx_xx@xx_xx|xx_xx/G:5_1%0_xx_xx/H:xx_xx/I:xx-xx@xx+xx&xx-xx|xx+xx/J:2_11/K:1+2-11"

Grapheme-to-phoeneme (G2P)

import jpreprocess

j = jpreprocess.jpreprocess()

assert j.g2p("おはようございます") == "o h a y o o g o z a i m a s U"
assert j.g2p("おはようございます", kana=True) == "オハヨーゴザイマス"

Copyrights

Please see README.md.

License

BSD-3-Clause