feat: 469/470 (99.8%) NeMo coverage

- electronic: exclude "dot" as email username first token
- money: reject singular "one" with plural currency ("one dollars")
- telephone: add credit card 4-6-4/4-6-5 formats with optional country code
- telephone: exclude "a" as serial first char to avoid "a thirty six" -> "a36"
- punctuation: add Punctuation class, split punct from words ("twenty!" -> "20 !")

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

Author: pengzhendong

itn/english/inverse_normalizer.py

@@ -24,6 +24,7 @@
 from itn.english.rules.measure import Measure
 from itn.english.rules.money import Money
 from itn.english.rules.ordinal import Ordinal
+from itn.english.rules.punctuation import Punctuation
 from itn.english.rules.telephone import Telephone
 from itn.english.rules.time import Time
 from itn.english.rules.whitelist import Whitelist
@@ -52,8 +53,9 @@ def build_tagger_and_verbalizer(self):
         whitelist = Whitelist()
         word = Word()
         char = Char()
+        punctuation = Punctuation()
 
-        tagger = (
+        classify = (
             add_weight(date.tagger, 1.09)
             | add_weight(time.tagger, 1.1)
             | add_weight(measure.tagger, 1.1)
@@ -68,7 +70,8 @@ def build_tagger_and_verbalizer(self):
             | add_weight(char.tagger, 100)
         ).optimize()
 
-        token = tagger
+        punct = add_weight(punctuation.tagger, 1.1)
+        token = closure(punct + delete(" ").ques) + classify + closure(delete(" ").ques + punct)
         graph = token + closure(self.DELETE_EXTRA_SPACE + token)
         self.tagger = delete(" ").star + graph + delete(" ").star
 
@@ -85,6 +88,7 @@ def build_tagger_and_verbalizer(self):
             | whitelist.verbalizer
             | word.verbalizer
             | char.verbalizer
+            | punctuation.verbalizer
         ).optimize()
 
         self.verbalizer = (verbalizer + self.INSERT_SPACE).star @ self.build_rule(

itn/english/rules/electronic.py

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from pynini import closure, cross, invert, string_file
+from pynini import accep, closure, cross, difference, invert, string_file
 from pynini.lib.pynutil import add_weight, delete, insert
 
 from tn.processor import Processor
@@ -35,7 +35,8 @@ def build_tagger(self):
         char = self.ALPHA | digit | zero
         word = add_weight(closure(self.ALPHA, 2), 0.1)
         token = char | symbols | word
-        component = token + closure(ds + token)
+        first_token = char | difference(word, accep("dot"))
+        component = first_token + closure(ds + token)
 
         dot = cross("dot", ".")
         domain = component + (ds + dot + ds + component).plus

itn/english/rules/money.py

@@ -37,14 +37,15 @@ def build_tagger(self):
         ds = delete(" ")
 
         currency_labels = load_labels(get_abs_path("../itn/english/data/currency.tsv"))
-        currency_pairs = []
-        for symbol, name in currency_labels:
-            currency_pairs.append((name, symbol))
+        singular_pairs = [(name, symbol) for symbol, name in currency_labels]
+        plural_pairs = []
+        for name, symbol in singular_pairs:
             if name.endswith("s"):
-                currency_pairs.append((name + "es", symbol))
+                plural_pairs.append((name + "es", symbol))
             else:
-                currency_pairs.append((name + "s", symbol))
-        currency = union(*[cross(name, symbol) for name, symbol in currency_pairs]).optimize()
+                plural_pairs.append((name + "s", symbol))
+        currency_singular = union(*[cross(name, symbol) for name, symbol in singular_pairs]).optimize()
+        currency_plural = union(*[cross(name, symbol) for name, symbol in singular_pairs + plural_pairs]).optimize()
 
         cent = cross("cent", "") | cross("cents", "")
         magnitudes = load_labels(get_abs_path("../itn/english/data/magnitudes.tsv"))
@@ -57,15 +58,23 @@ def build_tagger(self):
             compose(cardinal_graph, self.DIGIT ** 3),
         )
         cardinal_with_hundred = cardinal_graph | with_hundred
+        not_one = self.DIGIT ** (2, ...) | (self.DIGIT - accep("1"))
+        cardinal_plural = compose(cardinal_with_hundred, not_one)
+        # "one dollar" (singular) vs "two dollars" (plural)
+        one = cross("one", "1")
         integer_graph = (
-            insert('value: "') + cardinal_with_hundred + insert('"')
-            + ds + insert(' currency: "') + currency + insert('"')
+            insert('value: "') + cardinal_plural + insert('"')
+            + ds + insert(' currency: "') + currency_plural + insert('"')
+        )
+        integer_graph |= (
+            insert('value: "') + one + insert('"')
+            + ds + insert(' currency: "') + currency_singular + insert('"')
         )
         # "fifty million dollars" / "four hundred billion won"
         quantity_graph = (
             insert('value: "') + cardinal_small + insert('"')
             + ds + insert(' quantity: "') + magnitude + insert('"')
-            + ds + insert(' currency: "') + currency + insert('"')
+            + ds + insert(' currency: "') + currency_plural + insert('"')
         )
         # "two point five billion dollars"
         digit = string_file(get_abs_path("../itn/english/data/numbers/digit.tsv"))
@@ -76,38 +85,38 @@ def build_tagger(self):
             insert('value: "') + cardinal_graph + insert(".")
             + ds + delete("point") + ds + frac + insert('"')
             + ds + insert(' quantity: "') + magnitude + insert('"')
-            + ds + insert(' currency: "') + currency + insert('"')
+            + ds + insert(' currency: "') + currency_plural + insert('"')
         )
         # "twenty point five o six dollars" (decimal without quantity)
         decimal_graph = (
             insert('value: "') + cardinal_graph + insert(".")
             + ds + delete("point") + ds + frac + insert('"')
-            + ds + insert(' currency: "') + currency + insert('"')
+            + ds + insert(' currency: "') + currency_plural + insert('"')
         )
         # "point five o six dollars"
         decimal_no_int = (
             insert('value: ".') + delete("point") + ds + frac + insert('"')
-            + ds + insert(' currency: "') + currency + insert('"')
+            + ds + insert(' currency: "') + currency_plural + insert('"')
         )
         # "one fifty five dollars" => $155 (missing "hundred")
         with_hundred = (
             insert('value: "') + cardinal_small + insert('"')
-            + ds + insert(' currency: "') + currency + insert('"')
+            + ds + insert(' currency: "') + currency_plural + insert('"')
         )
 
         # cents
         cents_graph = union(*[cross(_num_to_word(x), f"{x:02d}") for x in range(1, 100) if _num_to_word(x)])
         with_cents = (
             insert('value: "') + cardinal_graph + insert('"')
-            + ds + insert(' currency: "') + currency + insert('"')
+            + ds + insert(' currency: "') + currency_plural + insert('"')
             + ds + (delete("and") + ds).ques
             + insert(' decimal: "') + cents_graph + insert('"')
             + ds + cent
         )
         # "seventy five dollars sixty three" (no "cents" word)
         dollars_amount = (
             insert('value: "') + cardinal_graph + insert('"')
-            + ds + insert(' currency: "') + currency + insert('"')
+            + ds + insert(' currency: "') + currency_plural + insert('"')
             + ds + insert(' decimal: "') + cents_graph + insert('"')
         )
         cents_only = (

itn/english/rules/punctuation.py

@@ -0,0 +1,31 @@
+# Copyright (c) 2026 Zhendong Peng (pzd17@tsinghua.org.cn)
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from pynini import union
+from pynini.lib.pynutil import insert
+
+from tn.processor import Processor
+
+
+class Punctuation(Processor):
+
+    def __init__(self):
+        super().__init__(name="punctuation", ordertype="itn")
+        self.build_tagger()
+        self.build_verbalizer()
+
+    def build_tagger(self):
+        punct = union(*"!#$%&'()*+,-./:;<=>?@^_`{|}~")
+        tagger = insert('value: "') + punct + insert('"')
+        self.tagger = self.add_tokens(tagger)

itn/english/rules/telephone.py

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from pynini import closure, cross, string_file, union
+from pynini import closure, cross, difference, string_file, union
 from pynini.lib.pynutil import add_weight, delete, insert
 
 from itn.english.rules.cardinal import Cardinal
@@ -82,17 +82,28 @@ def build_tagger(self):
         ip = ip_token + (cross(" dot ", ".") + ip_token) ** 3
         graph |= insert('number_part: "') + add_weight(ip, -0.001) + insert('"')
 
-        # credit card: XXXX XXXX XXXX XXXX or XXXX XXXXXX XXXXX
-        cc = seq @ (
-            self.DIGIT ** 4 + insert(" ") + self.DIGIT ** 4
-            + insert(" ") + self.DIGIT ** 4 + insert(" ") + self.DIGIT ** 4
+        # credit card: 4-4-4-4 (16), 4-6-4 (14), 4-6-5 (15)
+        space = insert(" ")
+        D = self.DIGIT
+        cc_format = (
+            D ** 4 + space + D ** 4 + space + D ** 4 + space + D ** 4
+            | D ** 4 + space + D ** 6 + space + D ** 4
+            | D ** 4 + space + D ** 6 + space + D ** 5
         )
-        graph |= insert('number_part: "') + cc + insert('"')
+        cc = seq @ cc_format
+        graph |= optional_cc + insert('number_part: "') + cc + insert('"')
 
         # serial: mixed alpha+digits, at least one digit, length >= 3
-        serial_char = add_weight(single, 0.001) | add_weight(two_digit, -0.001) | self.ALPHA
-        serial = serial_char + closure(ds + serial_char, 1)
-        serial = serial @ (closure(self.ALPHA | self.DIGIT) + self.DIGIT + closure(self.ALPHA | self.DIGIT))
+        # Exclude "a" as first char to avoid "a thirty six" -> "a36"
+        not_a = difference(self.ALPHA, union("a", "A"))
+        serial_digit = single | add_weight(two_digit, -0.002)
+        serial_char = serial_digit | self.ALPHA
+        seq1 = (not_a | serial_digit) + closure(ds + serial_char, 2)
+        seq1 |= serial_char + closure(ds + (single | self.ALPHA), 2)
+        seq2 = self.ALPHA + closure(ds + self.ALPHA, 1) + closure(ds + two_digit, 1)
+        seq2 |= not_a + closure(ds + two_digit, 1)
+        seq2 |= two_digit + closure(ds + two_digit, 1) + closure(ds + self.ALPHA, 1)
+        serial = (seq1 | seq2) @ (closure(self.ALPHA | D) + D + closure(self.ALPHA | D))
         graph |= insert('number_part: "') + add_weight(serial, 2.0) + insert('"')
 
         self.tagger = self.add_tokens(graph)

itn/english/test/data/en_electronic.txt

@@ -22,3 +22,4 @@ n vidia dot com => nvidia.com
 abc at gmail dot com => abc@gmail.com
 athreed at gmail dot com => athreed@gmail.com
 kore dot ai => kore.ai
+dot three at g mail dot com => dot 3@gmail.com

itn/english/test/data/en_money.txt

@@ -49,3 +49,4 @@ one fifty five dollars => $155
 fifteen hundred dollars => $1500
 ninety nine hundred dollars => $9900
 ninety nine hundred and fifteen dollars and one cent => $9915.01
+one dollars => one dollars

itn/english/test/data/en_telephone.txt

@@ -18,3 +18,6 @@ five w k r a three one => 5wkra31
 x eighty six => x86
 x three eighty six => x386
 r t x forty fifty t i => RTX 4050ti
+four three two double seven three two one four three two one four three double zero five => 432 7732 143214 3005
+a thirty six => a 36
+a ten eighty p display => a 1080p display

itn/english/test/data/en_word.txt

@@ -3,6 +3,7 @@
 , one , two , three , four => , one , two , three , four
 e s three => es3
 yahoo! => yahoo!
+twenty! => 20 !
 x => x
 — => —
 aaa => aaa