|
| 1 | +# Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved. |
| 2 | +# |
| 3 | +# Licensed under the Apache License, Version 2.0 (the "License"); |
| 4 | +# you may not use this file except in compliance with the License. |
| 5 | +# You may obtain a copy of the License at |
| 6 | +# |
| 7 | +# http://www.apache.org/licenses/LICENSE-2.0 |
| 8 | +# |
| 9 | +# Unless required by applicable law or agreed to in writing, software |
| 10 | +# distributed under the License is distributed on an "AS IS" BASIS, |
| 11 | +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 12 | +# See the License for the specific language governing permissions and |
| 13 | +# limitations under the License. |
| 14 | + |
| 15 | +import pynini |
| 16 | +from pynini.lib import pynutil |
| 17 | + |
| 18 | +from tn.processor import Processor |
| 19 | +from tn.utils import get_abs_path, load_labels, augment_labels_with_punct_at_end |
| 20 | +from tn.english.rules.cardinal import Cardinal |
| 21 | + |
| 22 | + |
| 23 | +class Time(Processor): |
| 24 | + |
| 25 | + def __init__(self, deterministic: bool = False): |
| 26 | + """ |
| 27 | + Args: |
| 28 | + deterministic: if True will provide a single transduction option, |
| 29 | + for False multiple transduction are generated (used for audio-based normalization) |
| 30 | + """ |
| 31 | + super().__init__('time', ordertype="en_tn") |
| 32 | + self.deterministic = deterministic |
| 33 | + self.build_tagger() |
| 34 | + self.build_verbalizer() |
| 35 | + |
| 36 | + def build_tagger(self): |
| 37 | + """ |
| 38 | + Finite state transducer for classifying time, e.g. |
| 39 | + 12:30 a.m. est -> time { hours: "twelve" minutes: "thirty" suffix: "a m" zone: "e s t" } |
| 40 | + 2.30 a.m. -> time { hours: "two" minutes: "thirty" suffix: "a m" } |
| 41 | + 02.30 a.m. -> time { hours: "two" minutes: "thirty" suffix: "a m" } |
| 42 | + 2.00 a.m. -> time { hours: "two" suffix: "a m" } |
| 43 | + 2 a.m. -> time { hours: "two" suffix: "a m" } |
| 44 | + 02:00 -> time { hours: "two" } |
| 45 | + 2:00 -> time { hours: "two" } |
| 46 | + 10:00:05 a.m. -> time { hours: "ten" minutes: "zero" seconds: "five" suffix: "a m" } |
| 47 | + """ |
| 48 | + suffix_labels = load_labels( |
| 49 | + get_abs_path("english/data/time/suffix.tsv")) |
| 50 | + suffix_labels.extend(augment_labels_with_punct_at_end(suffix_labels)) |
| 51 | + suffix_graph = pynini.string_map(suffix_labels) |
| 52 | + |
| 53 | + time_zone_graph = pynini.string_file( |
| 54 | + get_abs_path("english/data/time/zone.tsv")) |
| 55 | + |
| 56 | + # only used for < 1000 thousand -> 0 weight |
| 57 | + cardinal = Cardinal(self.deterministic).graph |
| 58 | + |
| 59 | + labels_hour = [str(x) for x in range(0, 24)] |
| 60 | + labels_minute_single = [str(x) for x in range(1, 10)] |
| 61 | + labels_minute_double = [str(x) for x in range(10, 60)] |
| 62 | + |
| 63 | + delete_leading_zero_to_double_digit = (self.DIGIT + self.DIGIT) | ( |
| 64 | + pynini.closure(pynutil.delete("0"), 0, 1) + self.DIGIT) |
| 65 | + |
| 66 | + graph_hour = delete_leading_zero_to_double_digit @ pynini.union( |
| 67 | + *labels_hour) @ cardinal |
| 68 | + |
| 69 | + graph_minute_single = pynini.union(*labels_minute_single) @ cardinal |
| 70 | + graph_minute_double = pynini.union(*labels_minute_double) @ cardinal |
| 71 | + |
| 72 | + final_graph_hour = pynutil.insert( |
| 73 | + "hours: \"") + graph_hour + pynutil.insert("\"") |
| 74 | + final_graph_minute = ( |
| 75 | + pynutil.insert("minutes: \"") + |
| 76 | + (pynini.cross("0", "o") + self.INSERT_SPACE + graph_minute_single |
| 77 | + | graph_minute_double) + pynutil.insert("\"")) |
| 78 | + final_graph_second = ( |
| 79 | + pynutil.insert("seconds: \"") + |
| 80 | + (pynini.cross("0", "o") + self.INSERT_SPACE + graph_minute_single |
| 81 | + | graph_minute_double) + pynutil.insert("\"")) |
| 82 | + final_suffix = pynutil.insert( |
| 83 | + "suffix: \"") + suffix_graph + pynutil.insert("\"") |
| 84 | + final_suffix_optional = pynini.closure( |
| 85 | + self.DELETE_SPACE + self.INSERT_SPACE + final_suffix, 0, 1) |
| 86 | + final_time_zone_optional = pynini.closure( |
| 87 | + self.DELETE_SPACE + self.INSERT_SPACE + |
| 88 | + pynutil.insert("zone: \"") + time_zone_graph + |
| 89 | + pynutil.insert("\""), |
| 90 | + 0, |
| 91 | + 1, |
| 92 | + ) |
| 93 | + |
| 94 | + # 2:30 pm, 02:30, 2:00 |
| 95 | + graph_hm = ( |
| 96 | + final_graph_hour + pynutil.delete(":") + |
| 97 | + (pynutil.delete("00") | self.INSERT_SPACE + final_graph_minute) + |
| 98 | + final_suffix_optional + final_time_zone_optional) |
| 99 | + |
| 100 | + # 10:30:05 pm, |
| 101 | + graph_hms = (final_graph_hour + pynutil.delete(":") + |
| 102 | + (pynini.cross("00", " minutes: \"zero\"") |
| 103 | + | self.INSERT_SPACE + final_graph_minute) + |
| 104 | + pynutil.delete(":") + |
| 105 | + (pynini.cross("00", " seconds: \"zero\"") |
| 106 | + | self.INSERT_SPACE + final_graph_second) + |
| 107 | + final_suffix_optional + final_time_zone_optional) |
| 108 | + |
| 109 | + # 2.xx pm/am |
| 110 | + graph_hm2 = ( |
| 111 | + final_graph_hour + pynutil.delete(".") + |
| 112 | + (pynutil.delete("00") | self.INSERT_SPACE + final_graph_minute) + |
| 113 | + self.DELETE_SPACE + self.INSERT_SPACE + final_suffix + |
| 114 | + final_time_zone_optional) |
| 115 | + # 2 pm est |
| 116 | + graph_h = final_graph_hour + self.DELETE_SPACE + self.INSERT_SPACE + final_suffix + final_time_zone_optional |
| 117 | + final_graph = (graph_hm | graph_h | graph_hm2 | graph_hms).optimize() |
| 118 | + |
| 119 | + final_graph = self.add_tokens(final_graph) |
| 120 | + self.tagger = final_graph.optimize() |
| 121 | + |
| 122 | + def build_verbalizer(self): |
| 123 | + """ |
| 124 | + Finite state transducer for verbalizing time, e.g. |
| 125 | + time { hours: "twelve" minutes: "thirty" suffix: "a m" zone: "e s t" } -> twelve thirty a m e s t |
| 126 | + time { hours: "twelve" } -> twelve o'clock |
| 127 | + """ |
| 128 | + hour = (pynutil.delete("hours:") + self.DELETE_SPACE + |
| 129 | + pynutil.delete("\"") + pynini.closure(self.NOT_QUOTE, 1) + |
| 130 | + pynutil.delete("\"")) |
| 131 | + minute = (pynutil.delete("minutes:") + self.DELETE_SPACE + |
| 132 | + pynutil.delete("\"") + pynini.closure(self.NOT_QUOTE, 1) + |
| 133 | + pynutil.delete("\"")) |
| 134 | + suffix = (pynutil.delete("suffix:") + self.DELETE_SPACE + |
| 135 | + pynutil.delete("\"") + pynini.closure(self.NOT_QUOTE, 1) + |
| 136 | + pynutil.delete("\"")) |
| 137 | + optional_suffix = pynini.closure( |
| 138 | + self.DELETE_SPACE + self.INSERT_SPACE + suffix, 0, 1) |
| 139 | + zone = (pynutil.delete("zone:") + self.DELETE_SPACE + |
| 140 | + pynutil.delete("\"") + pynini.closure(self.NOT_QUOTE, 1) + |
| 141 | + pynutil.delete("\"")) |
| 142 | + optional_zone = pynini.closure( |
| 143 | + self.DELETE_SPACE + self.INSERT_SPACE + zone, 0, 1) |
| 144 | + second = (pynutil.delete("seconds:") + self.DELETE_SPACE + |
| 145 | + pynutil.delete("\"") + pynini.closure(self.NOT_QUOTE, 1) + |
| 146 | + pynutil.delete("\"")) |
| 147 | + graph_hms = (hour + pynutil.insert(" hours ") + self.DELETE_SPACE + |
| 148 | + minute + pynutil.insert(" minutes and ") + |
| 149 | + self.DELETE_SPACE + second + pynutil.insert(" seconds") + |
| 150 | + optional_suffix + optional_zone) |
| 151 | + graph_hms @= pynini.cdrewrite( |
| 152 | + pynutil.delete("o ") |
| 153 | + | pynini.cross("one minutes", "one minute") |
| 154 | + | pynini.cross("one seconds", "one second") |
| 155 | + | pynini.cross("one hours", "one hour"), |
| 156 | + pynini.union(" ", "[BOS]"), |
| 157 | + "", |
| 158 | + pynini.closure(self.VCHAR), |
| 159 | + ) |
| 160 | + graph = hour + self.DELETE_SPACE + self.INSERT_SPACE + minute + optional_suffix + optional_zone |
| 161 | + graph |= hour + self.INSERT_SPACE + pynutil.insert( |
| 162 | + "o'clock") + optional_zone |
| 163 | + graph |= hour + self.DELETE_SPACE + self.INSERT_SPACE + suffix + optional_zone |
| 164 | + graph |= graph_hms |
| 165 | + delete_tokens = self.delete_tokens(graph) |
| 166 | + self.verbalizer = delete_tokens.optimize() |
0 commit comments