bootphon/phonemizer

Two phones in Arabic are mapped to the same symbol

Closed this issue · 1 comments

LamOne1 commented

I noticed that "ظ" (ðˤ) and "ذ" (ð) are both mapped to "ð"

from phonemizer.backend import BACKENDS
g2p = BACKENDS['espeak'](language='ar', words_mismatch='warn', preserve_punctuation=False)
g2p.phonemize(["ظ"], separator=Separator(phone=None, word=' ', syllable='|'), strip=True)

# -----> ðaaʔ

g2p.phonemize(["ذ"], separator=Separator(phone=None, word=' ', syllable='|'), strip=True)

# -----> ðaal

packages:
phonemizer 3.2.1
python 3.8.19

mmmaat commented

Hi, this is related to the espeak backend, not phonemizer. Nothing we can do here. You may want to open an issue in the espeak-ng repo.

$ espeak-ng -q -x --ipa -v ar "ظ"
ðˈaaʔ
$ espeak-ng -q -x --ipa -v ar "ذ"
ðˈaal
  • 👍1