Vietnamese TN - Fraction

nemo_text_processing/text_normalization/vi/data/fraction/__init__.py

@@ -0,0 +1,13 @@
+# Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES.  All rights reserved.
+#
+# 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.

nemo_text_processing/text_normalization/vi/data/fraction/denominator_exceptions.tsv

@@ -0,0 +1 @@
+4	tư

nemo_text_processing/text_normalization/vi/taggers/fraction.py

@@ -0,0 +1,67 @@
+# Copyright (c) 2023, NVIDIA CORPORATION.  All rights reserved.
+#
+# 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.
+
+import pynini
+from pynini.lib import pynutil
+
+from nemo_text_processing.text_normalization.en.graph_utils import GraphFst
+from nemo_text_processing.text_normalization.vi.taggers.cardinal import CardinalFst
+from nemo_text_processing.text_normalization.vi.utils import get_abs_path, load_labels
+
+
+class FractionFst(GraphFst):
+    """
+    Finite state transducer for classifying Vietnamese fraction numbers, e.g.
+        23 1/5 -> fraction { integer_part: "hai mươi ba" numerator: "một" denominator: "năm" }
+        3/9 -> fraction { numerator: "ba" denominator: "chín" }
+        1/4 -> fraction { numerator: "một" denominator: "tư" }
+
+    Args:
+        cardinal: CardinalFst for converting numbers to Vietnamese words
+        deterministic: if True will provide a single transduction option,
+            for False multiple options (used for audio-based normalization)
+    """
+
+    def __init__(self, cardinal: CardinalFst, deterministic: bool = True):
+        super().__init__(name="fraction", kind="classify", deterministic=deterministic)
+
+        cardinal_graph = cardinal.graph
+        digit = pynini.union(*[str(i) for i in range(10)])
+        number = pynini.closure(digit, 1)
+
+        denominator_exceptions = {
+            row[0]: row[1] for row in load_labels(get_abs_path("data/fraction/denominator_exceptions.tsv"))
+        }
+
+        denominator_exception_patterns = [pynini.cross(k, v) for k, v in denominator_exceptions.items()]
+        denominator_exception_graph = (
+            pynini.union(*denominator_exception_patterns) if denominator_exception_patterns else None
+        )
+        denominator_graph = (
+            pynini.union(denominator_exception_graph, cardinal_graph)
+            if denominator_exception_graph
+            else cardinal_graph
+        )
+
+        numerator = (
+            pynutil.insert("numerator: \"") + (number @ cardinal_graph) + pynutil.insert("\" ") + pynutil.delete("/")
+        )
+        denominator = pynutil.insert("denominator: \"") + (number @ denominator_graph) + pynutil.insert("\"")
+        integer_part = pynutil.insert("integer_part: \"") + (number @ cardinal_graph) + pynutil.insert("\" ")
+
+        simple_fraction = numerator + denominator
+        mixed_fraction = integer_part + pynutil.delete(" ") + numerator + denominator
+        optional_graph_negative = (pynutil.insert("negative: ") + pynini.cross("-", "\"true\" ")).ques
+
+        self.fst = self.add_tokens(optional_graph_negative + (simple_fraction | mixed_fraction)).optimize()

nemo_text_processing/text_normalization/vi/taggers/tokenize_and_classify.py

@@ -26,6 +26,7 @@
 )
 from nemo_text_processing.text_normalization.vi.taggers.cardinal import CardinalFst
 from nemo_text_processing.text_normalization.vi.taggers.decimal import DecimalFst
+from nemo_text_processing.text_normalization.vi.taggers.fraction import FractionFst
 from nemo_text_processing.text_normalization.vi.taggers.ordinal import OrdinalFst
 from nemo_text_processing.text_normalization.vi.taggers.punctuation import PunctuationFst
 from nemo_text_processing.text_normalization.vi.taggers.whitelist import WhiteListFst
@@ -86,11 +87,17 @@ def __init__(
             decimal_graph = decimal.fst
             logger.debug(f"decimal: {time.time() - start_time: .2f}s -- {decimal_graph.num_states()} nodes")
 
+            start_time = time.time()
+            fraction = FractionFst(cardinal=cardinal, deterministic=deterministic)
+            fraction_graph = fraction.fst
+            logger.debug(f"fraction: {time.time() - start_time: .2f}s -- {fraction_graph.num_states()} nodes")
+
             classify = (
                 pynutil.add_weight(whitelist_graph, 0.8)
                 | pynutil.add_weight(ordinal_graph, 0.81)
                 | pynutil.add_weight(decimal_graph, 0.85)
                 | pynutil.add_weight(cardinal_graph, 0.9)
+                | pynutil.add_weight(fraction_graph, 1.0)
                 | pynutil.add_weight(word_graph, 100)
             )
             punct = pynutil.insert("tokens { ") + pynutil.add_weight(punct_graph, weight=2.1) + pynutil.insert(" }")

nemo_text_processing/text_normalization/vi/verbalizers/fraction.py

@@ -0,0 +1,53 @@
+# Copyright (c) 2023, NVIDIA CORPORATION.  All rights reserved.
+#
+# 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.
+
+import pynini
+from pynini.lib import pynutil
+
+from nemo_text_processing.text_normalization.en.graph_utils import NEMO_NOT_QUOTE, GraphFst, delete_space
+
+
+class FractionFst(GraphFst):
+    """
+    Finite state transducer for verbalizing Vietnamese fraction numbers, e.g.
+        fraction { negative: "true" integer_part: "hai mươi ba" numerator: "một" denominator: "năm" } -> âm hai mươi ba và một phần năm
+        fraction { numerator: "ba" denominator: "chín" } -> ba phần chín
+        fraction { integer_part: "một trăm" numerator: "hai" denominator: "ba" } -> một trăm và hai phần ba
+
+    Args:
+        deterministic: if True will provide a single transduction option,
+            for False multiple options (used for audio-based normalization)
+    """
+
+    def __init__(self, deterministic: bool = True):
+        super().__init__(name="fraction", kind="verbalize", deterministic=deterministic)
+
+        optional_sign = pynini.cross("negative: \"true\"", "âm ")
+        if not deterministic:
+            optional_sign |= pynini.cross("negative: \"true\"", "trừ ")
+        optional_sign = pynini.closure(optional_sign + delete_space, 0, 1)
+
+        part = pynini.closure(NEMO_NOT_QUOTE)
+        delete_quotes = delete_space + pynutil.delete("\"") + part + pynutil.delete("\"")
+
+        integer_tagged = pynutil.delete("integer_part:") + delete_quotes
+        numerator_tagged = pynutil.delete("numerator:") + delete_quotes
+        denominator_tagged = pynutil.delete("denominator:") + delete_quotes
+
+        fraction_part = numerator_tagged + delete_space + pynutil.insert(" phần ") + denominator_tagged
+
+        simple_fraction = fraction_part
+        mixed_fraction = integer_tagged + delete_space + pynutil.insert(" và ") + fraction_part
+
+        self.fst = self.delete_tokens(optional_sign + (simple_fraction | mixed_fraction)).optimize()

nemo_text_processing/text_normalization/vi/verbalizers/verbalize.py

@@ -16,6 +16,7 @@
 from nemo_text_processing.text_normalization.en.verbalizers.word import WordFst
 from nemo_text_processing.text_normalization.vi.verbalizers.cardinal import CardinalFst
 from nemo_text_processing.text_normalization.vi.verbalizers.decimal import DecimalFst
+from nemo_text_processing.text_normalization.vi.verbalizers.fraction import FractionFst
 from nemo_text_processing.text_normalization.vi.verbalizers.ordinal import OrdinalFst
 from nemo_text_processing.text_normalization.vi.verbalizers.whitelist import WhiteListFst
 
@@ -40,7 +41,10 @@ def __init__(self, deterministic: bool = True):
         decimal = DecimalFst(cardinal=cardinal, deterministic=deterministic)
         decimal_graph = decimal.fst
 
+        fraction = FractionFst(deterministic=deterministic)
+        fraction_graph = fraction.fst
+
         # Combine all verbalizers
-        graph = cardinal_graph | whitelist_graph | word_graph | ordinal_graph | decimal_graph
+        graph = cardinal_graph | whitelist_graph | word_graph | ordinal_graph | decimal_graph | fraction_graph
 
         self.fst = graph

tests/nemo_text_processing/vi/data_text_normalization/test_cases_fraction.txt

@@ -0,0 +1,13 @@
+1/2~một phần hai
+4/9~bốn phần chín
+9/4~chín phần tư
+1/4~một phần tư
+3/4~ba phần tư
+15/5~mười lăm phần năm
+1/3~một phần ba
+2/10~hai phần mười
+23 1/5~hai mươi ba và một phần năm
+-3/4~âm ba phần tư
+-12 1/4 nha~âm mười hai và một phần tư nha
+-5 2/3~âm năm và hai phần ba
+5 1/2~năm và một phần hai

tests/nemo_text_processing/vi/test_fraction.py

@@ -12,32 +12,49 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-
+# pytest tests/nemo_text_processing/vi/test_fraction.py --cpu --cache-clear
 import pytest
 from parameterized import parameterized
 
-from ..utils import CACHE_DIR, parse_test_case_file
-
-try:
-    from nemo_text_processing.inverse_text_normalization.inverse_normalize import InverseNormalizer
+from nemo_text_processing.inverse_text_normalization.inverse_normalize import InverseNormalizer
+from nemo_text_processing.text_normalization.normalize import Normalizer
+from nemo_text_processing.text_normalization.normalize_with_audio import NormalizerWithAudio
 
-    PYNINI_AVAILABLE = True
-except (ImportError, ModuleNotFoundError):
-    PYNINI_AVAILABLE = False
+from ..utils import CACHE_DIR, RUN_AUDIO_BASED_TESTS, parse_test_case_file
 
 
 class TestFraction:
-    inverse_normalizer = (
-        InverseNormalizer(lang='vi', cache_dir=CACHE_DIR, overwrite_cache=False) if PYNINI_AVAILABLE else None
-    )
+
+    inverse_normalizer = InverseNormalizer(lang='vi', cache_dir=CACHE_DIR, overwrite_cache=False)
 
     @parameterized.expand(parse_test_case_file('vi/data_inverse_text_normalization/test_cases_fraction.txt'))
-    @pytest.mark.skipif(
-        not PYNINI_AVAILABLE,
-        reason="`pynini` not installed, please install via nemo_text_processing/pynini_install.sh",
-    )
     @pytest.mark.run_only_on('CPU')
     @pytest.mark.unit
     def test_denorm(self, test_input, expected):
         pred = self.inverse_normalizer.inverse_normalize(test_input, verbose=False)
         assert pred == expected
+
+    normalizer = Normalizer(
+        input_case='cased', lang='vi', cache_dir=CACHE_DIR, overwrite_cache=False, post_process=True
+    )
+
+    normalizer_with_audio = (
+        NormalizerWithAudio(input_case='cased', lang='vi', cache_dir=CACHE_DIR, overwrite_cache=False)
+        if CACHE_DIR and RUN_AUDIO_BASED_TESTS
+        else None
+    )
+
+    @parameterized.expand(parse_test_case_file('vi/data_text_normalization/test_cases_fraction.txt'))
+    @pytest.mark.run_only_on('CPU')
+    @pytest.mark.unit
+    def test_norm(self, test_input, expected):
+        pred = self.normalizer.normalize(test_input, verbose=False, punct_post_process=False)
+        assert pred == expected, f"input: {test_input}"
+
+        if self.normalizer_with_audio:
+            pred_non_deterministic = self.normalizer_with_audio.normalize(
+                test_input,
+                n_tagged=30,
+                punct_post_process=False,
+            )
+            assert expected in pred_non_deterministic, f"input: {test_input}"

tests/nemo_text_processing/vi/test_sparrowhawk_normalization.sh

@@ -43,10 +43,10 @@ testTNOrdinal() {
   runtest $input
 }
 
-# testTNFraction() {
-#   input=$PROJECT_DIR/vi/data_text_normalization/test_cases_fraction.txt
-#   runtest $input
-# }
+testTNFraction() {
+  input=$PROJECT_DIR/vi/data_text_normalization/test_cases_fraction.txt
+  runtest $input
+}
 
 # testTNTime() {
 #   input=$PROJECT_DIR/vi/data_text_normalization/test_cases_time.txt