The punctuation mark is not separated correctly when using space as the phone separator in v0.3.1

[kan-bayashi]

Thank you for developing a great tool. This is very useful to develop our speech processing toolkit. I found unexpected behavior in v0.3.1.

Describe the bug The punctuation mark is not separated correctly when using space as the phone separator.

Phonemizer version

• 0.3.1

System

• OS information: Linux 3.10.0-1062.9.1.el7.x86_64 #1 SMP Mon Dec 2 08:31:54 EST 2019 x86_64
• python version: 3.9.7 (default, Sep 16 2021, 13:09:58) [GCC 7.5.0]

To reproduce

from phonemizer import phonemize
from phonemizer.separator import Separator

text = "Hello, world."
separator = Separator(
    word=None,
    syllable=None,
    phone=" ",
)
phn = phonemize(text, separator=separator, preserve_punctuation=True)
print(phn)

Expected behavior In phonemizer==0.3 (expected)

$ python test_phonemizer.py
h ə l oʊ , w ɜː l d .

In phonemizer==0.3.1 (seems wrong)

$ python test_phonemizer.py
h ə l oʊ, w ɜː l d.

[mmmaat]

in phonemizer-3.2.1this have been fixed by #119, the result is now h ə l oʊ ,w ɜː l d ., which is what is expected.