Spaces:
Sleeping
Sleeping
txya900619
commited on
Commit
•
fdeff58
1
Parent(s):
1280207
fix: ? ! can't parse to ipa error and add ignore_comma arg for future model
Browse files- ipa/ipa.py +10 -3
ipa/ipa.py
CHANGED
@@ -1,3 +1,6 @@
|
|
|
|
|
|
|
|
1 |
def can_form_string(x, symbol_dict):
|
2 |
def helper(x, symbol_dict, matched_parts):
|
3 |
if not x:
|
@@ -16,15 +19,19 @@ def can_form_string(x, symbol_dict):
|
|
16 |
return helper(x, symbol_dict, [])
|
17 |
|
18 |
|
19 |
-
def text_to_ipa(text, lang_tag, g2p):
|
20 |
ipa = []
|
21 |
-
|
|
|
|
|
|
|
22 |
|
23 |
print(words)
|
24 |
|
25 |
for word in words:
|
26 |
ipa_parts = ""
|
27 |
-
|
|
|
28 |
|
29 |
if result is False:
|
30 |
print(word)
|
|
|
1 |
+
import re
|
2 |
+
|
3 |
+
|
4 |
def can_form_string(x, symbol_dict):
|
5 |
def helper(x, symbol_dict, matched_parts):
|
6 |
if not x:
|
|
|
19 |
return helper(x, symbol_dict, [])
|
20 |
|
21 |
|
22 |
+
def text_to_ipa(text, lang_tag, g2p, ignore_comma=True):
|
23 |
ipa = []
|
24 |
+
text = text.lower()
|
25 |
+
text = re.sub(r"[.?!]", "", text)
|
26 |
+
text = text.replace("'", "’")
|
27 |
+
words = text.split() # change in future
|
28 |
|
29 |
print(words)
|
30 |
|
31 |
for word in words:
|
32 |
ipa_parts = ""
|
33 |
+
extended_g2p = {**g2p[lang_tag], ",": "" if ignore_comma else ","}
|
34 |
+
result, matched_parts = can_form_string(word, extended_g2p)
|
35 |
|
36 |
if result is False:
|
37 |
print(word)
|