-
Notifications
You must be signed in to change notification settings - Fork 4
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Showing
15 changed files
with
354 additions
and
32 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
70 changes: 70 additions & 0 deletions
70
nisaba/scripts/natural_translit/phonology/features/docs/language.md
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,70 @@ | ||
| # Languages | ||
|
|
||
| Feature family tree for the supported | ||
| [languages](https://github.com/google-research/nisaba/blob/main/nisaba/scripts/natural_translit/phonology/features/README.md#language). | ||
|
|
||
| <!-- AUTO-GENERATED INVENTORY STRING STARTS HERE --> | ||
|
|
||
| ```dot | ||
| graph { | ||
| ordering="out" | ||
| size = 12 | ||
| rankdir="LR" | ||
| ranksep="1, equally" | ||
| style="invis" | ||
| node [shape="plain"] | ||
| language [label="Language"] | ||
| dravidian [label="Dravidian"] | ||
| language -- dravidian | ||
| subgraph { | ||
| kn [label="Kannada"] | ||
| dravidian -- kn | ||
| te [label="Telugu"] | ||
| dravidian -- te | ||
| tamil_kota [label="Tamil-Kota"] | ||
| dravidian -- tamil_kota | ||
| subgraph { | ||
| ml [label="Malayalam"] | ||
| tamil_kota -- ml | ||
| ta [label="Tamil"] | ||
| tamil_kota -- ta | ||
| } | ||
| } | ||
| indo_european [label="Indo-European"] | ||
| language -- indo_european | ||
| subgraph { | ||
| indo_aryan [label="Indo-Aryan"] | ||
| indo_european -- indo_aryan | ||
| subgraph { | ||
| bn [label="Bengali"] | ||
| indo_aryan -- bn | ||
| gu [label="Gujarati"] | ||
| indo_aryan -- gu | ||
| hi [label="Hindi"] | ||
| indo_aryan -- hi | ||
| mr [label="Marathi"] | ||
| indo_aryan -- mr | ||
| pa [label="Panjabi"] | ||
| indo_aryan -- pa | ||
| } | ||
| germanic [label="Germanic"] | ||
| indo_european -- germanic | ||
| subgraph { | ||
| en [label="English"] | ||
| germanic -- en | ||
| } | ||
| } | ||
| mixed_family [label="Mixed Family Tags"] | ||
| language -- mixed_family | ||
| subgraph { | ||
| mul [label="Multiple Languages"] | ||
| mixed_family -- mul | ||
| und [label="Undetermined"] | ||
| mixed_family -- und | ||
| x_uni [label="Unified Multilingual"] | ||
| mixed_family -- x_uni | ||
| x_psa [label="Pan South Asian"] | ||
| mixed_family -- x_psa | ||
| } | ||
| } | ||
| ``` |
128 changes: 128 additions & 0 deletions
128
nisaba/scripts/natural_translit/phonology/features/language.py
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,128 @@ | ||
| # Copyright 2024 Nisaba Authors. | ||
| # | ||
| # 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. | ||
|
|
||
| """Language feature for PhonologicalSymbol.""" | ||
|
|
||
| from __future__ import annotations | ||
|
|
||
| from typing import Union | ||
|
|
||
| import pycountry | ||
|
|
||
| from nisaba.scripts.natural_translit.utils import feature as ft | ||
| from nisaba.scripts.natural_translit.utils import type_op as ty | ||
|
|
||
|
|
||
| class Language(ft.Feature): | ||
| """Language feature for Phon.""" | ||
|
|
||
| OR_NOTHING = Union['Language', ty.Nothing] | ||
|
|
||
| def __init__(self, alias: str, text: str, index: int): | ||
| super().__init__(alias, text) | ||
| self.index = index | ||
|
|
||
| @classmethod | ||
| def from_iso(cls, iso_code: str) -> Language: | ||
| if len(iso_code) == 2: | ||
| language = pycountry.languages.get(alpha_2=iso_code) | ||
| elif len(iso_code) == 3: | ||
| language = pycountry.languages.get(alpha_3=iso_code) | ||
| else: | ||
| raise ValueError(f'Invalid ISO code: {iso_code}') | ||
| return cls( | ||
| iso_code.lower(), | ||
| language.name, | ||
| list(pycountry.languages).index(language) | ||
| ) | ||
|
|
||
| class Inventory(ft.Feature.Inventory): | ||
| """Language feature inventory.""" | ||
|
|
||
| def __str__(self) -> str: | ||
| """Language family tree as markdown dot code block.""" | ||
| # Currently all languages are equidistant and have 1 distance from each | ||
| # other so there is no need for distance tables. If the subgroup steps | ||
| # are changed to reflect similarity of related languages, the distance | ||
| # tables should be appended below. | ||
| return self.language.visualize() | ||
|
|
||
|
|
||
| def _features() -> Language.Inventory: | ||
| """Language feature inventory. | ||
| Returns: | ||
| An inventory of languages. | ||
| The language inventory is organised by language families. The depth and the | ||
| granularity of the family hierarchy is pragmatic and can be updated as needed. | ||
| """ | ||
| f = ft.Feature | ||
| features = Language.Inventory( | ||
| 'languages', | ||
| f.Aspect( | ||
| f.equidistant( | ||
| 'language', | ||
| f.equidistant( | ||
| 'dravidian', | ||
| Language.from_iso('kn'), | ||
| Language.from_iso('te'), | ||
| f.equidistant( | ||
| ('tamil_kota', 'Tamil-Kota'), | ||
| Language.from_iso('ml'), | ||
| Language.from_iso('ta'), | ||
| ), | ||
| ), | ||
| f.equidistant( | ||
| ('indo_european', 'Indo-European'), | ||
| f.equidistant( | ||
| ('indo_aryan', 'Indo-Aryan'), | ||
| Language.from_iso('bn'), | ||
| Language.from_iso('gu'), | ||
| Language.from_iso('hi'), | ||
| Language.from_iso('mr'), | ||
| Language.from_iso('pa'), | ||
| ), | ||
| f.equidistant( | ||
| 'germanic', | ||
| Language.from_iso('en'), | ||
| ), | ||
| ), | ||
| f.equidistant( | ||
| ('mixed_family', 'Mixed Family Tags'), | ||
| # ISO 'mul' is a generic code for mixed-language datasets. | ||
| Language.from_iso('mul'), | ||
| Language.from_iso('und'), | ||
| # x_uni is a custom code for the unified phoneme inventory | ||
| # that will cover all IPA symbols and PHOIBLE segments. Index | ||
| # 8000 is reserved for the unified inventory. This will | ||
| # provide unique indices fol all phonemes as long as the | ||
| # number of pycountry languages is less than 8000 (currently | ||
| # 7874). If the number of pycounty languages exceeds 8000, we | ||
| # can change ReservedIndex for phoneme to 30_000_000 and | ||
| # LANG_PREFIX_MULTIPLIER to 1_000 to accommodate. | ||
| Language('x_uni', 'Unified Multilingual', 8_000), | ||
| # x_uni will potentially cover 2k to 6k segments, resulting in | ||
| # symbol indices from 3_800_001 up to 3_806_xxx. In order to | ||
| # avoid symbol index collision, we start other custom language | ||
| # indices from 9000. | ||
| Language('x_psa', 'Pan South Asian', 9_001), | ||
| ), | ||
| ), | ||
| ), | ||
| ) | ||
| return features | ||
|
|
||
|
|
||
| FEATURES = _features() |
33 changes: 33 additions & 0 deletions
33
nisaba/scripts/natural_translit/phonology/features/language_test.py
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,33 @@ | ||
| # Copyright 2024 Nisaba Authors. | ||
| # | ||
| # 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 absl.testing import absltest | ||
| from nisaba.scripts.natural_translit.phonology.features import language | ||
| from nisaba.scripts.natural_translit.utils import test_op | ||
|
|
||
| _LANG = language.FEATURES.language | ||
|
|
||
|
|
||
| class LanguageTest(test_op.TestCase): | ||
|
|
||
| def test_language(self): | ||
| self.assertEqual(_LANG.bn.index, 624) | ||
| self.assertEqual(_LANG.x_uni.index, 8_000) | ||
|
|
||
| def test_no_duplicate_indices(self): | ||
| index_list = [l.index for l in _LANG] | ||
| self.assertEqual(len(set(index_list)), len(index_list)) | ||
|
|
||
| if __name__ == '__main__': | ||
| absltest.main() |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
File renamed without changes.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.