From 1b9d44cf344cfa3edc086fff87b0a995f679193b Mon Sep 17 00:00:00 2001 From: Brandon Thomas Date: Mon, 4 Jul 2022 02:47:40 -0400 Subject: [PATCH] refactor --- tts/text/normalization/english_normalization.py | 17 ----------------- tts/text/pipeline/english_pipeline_v1.py | 10 +++++----- .../input_parsing/__init__.py | 0 .../input_parsing/parser.py | 0 .../input_parsing/test_parser.py | 2 +- .../normalization/__init__.py | 0 .../normalization/base_normalization.py | 0 .../normalization/common_cleaners.py | 0 .../normalization/english_normalization.py | 17 +++++++++++++++++ .../normalization/spanish_normalization.py | 2 +- .../normalization/test_english_normalization.py | 2 +- .../phonetization/__init__.py | 0 .../phonetization/base_phonetization.py | 2 +- .../phonetization/cmudict.py | 0 .../phonetization/english_phonetization.py | 6 +++--- .../phonetization/test_cmudict.py | 4 ++-- .../phonetization/test_english_phonetization.py | 8 ++++---- .../sequencing/__init__.py | 0 .../sequencing/english_symbols_v1.py | 2 +- .../sequencing/test_english_symbols_v1.py | 2 +- 20 files changed, 37 insertions(+), 37 deletions(-) delete mode 100644 tts/text/normalization/english_normalization.py rename tts/text/{ => pipeline_components}/input_parsing/__init__.py (100%) rename tts/text/{ => pipeline_components}/input_parsing/parser.py (100%) rename tts/text/{ => pipeline_components}/input_parsing/test_parser.py (93%) rename tts/text/{ => pipeline_components}/normalization/__init__.py (100%) rename tts/text/{ => pipeline_components}/normalization/base_normalization.py (100%) rename tts/text/{ => pipeline_components}/normalization/common_cleaners.py (100%) create mode 100644 tts/text/pipeline_components/normalization/english_normalization.py rename tts/text/{ => pipeline_components}/normalization/spanish_normalization.py (58%) rename tts/text/{ => pipeline_components}/normalization/test_english_normalization.py (81%) rename tts/text/{ => pipeline_components}/phonetization/__init__.py (100%) rename tts/text/{ => pipeline_components}/phonetization/base_phonetization.py (87%) rename tts/text/{ => pipeline_components}/phonetization/cmudict.py (100%) rename tts/text/{ => pipeline_components}/phonetization/english_phonetization.py (79%) rename tts/text/{ => pipeline_components}/phonetization/test_cmudict.py (72%) rename tts/text/{ => pipeline_components}/phonetization/test_english_phonetization.py (88%) rename tts/text/{ => pipeline_components}/sequencing/__init__.py (100%) rename tts/text/{ => pipeline_components}/sequencing/english_symbols_v1.py (96%) rename tts/text/{ => pipeline_components}/sequencing/test_english_symbols_v1.py (90%) diff --git a/tts/text/normalization/english_normalization.py b/tts/text/normalization/english_normalization.py deleted file mode 100644 index bf2c1e7..0000000 --- a/tts/text/normalization/english_normalization.py +++ /dev/null @@ -1,17 +0,0 @@ -from tts.text.normalization.base_normalization import BaseNormalization -from tts.text.normalization.common_cleaners import collapse_whitespace -from tts.text.normalization.common_cleaners import convert_to_ascii -from tts.text.normalization.common_cleaners import lowercase - -class EnglishNormalization(BaseNormalization): - - def normalize(self, raw_input_text: str) -> str: - text = convert_to_ascii(raw_input_text) - text = lowercase(text) - - # TODO: Expand abbreviations - # TODO: Expand numbers - - text = collapse_whitespace(text) - - return text diff --git a/tts/text/pipeline/english_pipeline_v1.py b/tts/text/pipeline/english_pipeline_v1.py index 8f67066..1d4f1c1 100644 --- a/tts/text/pipeline/english_pipeline_v1.py +++ b/tts/text/pipeline/english_pipeline_v1.py @@ -1,10 +1,10 @@ from typing import List -from tts.text.input_parsing.parser import ArpabetSequence, GraphemeSequence, ParsedSequence, parse_grapheme_and_arpabet_sequence -from tts.text.normalization.english_normalization import EnglishNormalization -from tts.text.phonetization.cmudict import DEFAULT_CMUDICT_PATH, CMUDict -from tts.text.phonetization.english_phonetization import EnglishPhonetization from tts.text.pipeline.base_pipeline import BasePipeline -from tts.text.sequencing.english_symbols_v1 import SPACE_INTEGER_ENCODING, arpabet_to_sequence, symbols_to_sequence +from tts.text.pipeline_components.input_parsing.parser import ArpabetSequence, GraphemeSequence, ParsedSequence, parse_grapheme_and_arpabet_sequence +from tts.text.pipeline_components.normalization.english_normalization import EnglishNormalization +from tts.text.pipeline_components.phonetization.cmudict import DEFAULT_CMUDICT_PATH, CMUDict +from tts.text.pipeline_components.phonetization.english_phonetization import EnglishPhonetization +from tts.text.pipeline_components.sequencing.english_symbols_v1 import SPACE_INTEGER_ENCODING, arpabet_to_sequence, symbols_to_sequence _ENGLISH_NORMALIZER = EnglishNormalization() diff --git a/tts/text/input_parsing/__init__.py b/tts/text/pipeline_components/input_parsing/__init__.py similarity index 100% rename from tts/text/input_parsing/__init__.py rename to tts/text/pipeline_components/input_parsing/__init__.py diff --git a/tts/text/input_parsing/parser.py b/tts/text/pipeline_components/input_parsing/parser.py similarity index 100% rename from tts/text/input_parsing/parser.py rename to tts/text/pipeline_components/input_parsing/parser.py diff --git a/tts/text/input_parsing/test_parser.py b/tts/text/pipeline_components/input_parsing/test_parser.py similarity index 93% rename from tts/text/input_parsing/test_parser.py rename to tts/text/pipeline_components/input_parsing/test_parser.py index 6545bc5..175b0d9 100644 --- a/tts/text/input_parsing/test_parser.py +++ b/tts/text/pipeline_components/input_parsing/test_parser.py @@ -1,4 +1,4 @@ -from tts.text.input_parsing.parser import ArpabetSequence, GraphemeSequence, parse_grapheme_and_arpabet_sequence +from tts.text.pipeline_components.input_parsing.parser import ArpabetSequence, GraphemeSequence, parse_grapheme_and_arpabet_sequence class TestParser: diff --git a/tts/text/normalization/__init__.py b/tts/text/pipeline_components/normalization/__init__.py similarity index 100% rename from tts/text/normalization/__init__.py rename to tts/text/pipeline_components/normalization/__init__.py diff --git a/tts/text/normalization/base_normalization.py b/tts/text/pipeline_components/normalization/base_normalization.py similarity index 100% rename from tts/text/normalization/base_normalization.py rename to tts/text/pipeline_components/normalization/base_normalization.py diff --git a/tts/text/normalization/common_cleaners.py b/tts/text/pipeline_components/normalization/common_cleaners.py similarity index 100% rename from tts/text/normalization/common_cleaners.py rename to tts/text/pipeline_components/normalization/common_cleaners.py diff --git a/tts/text/pipeline_components/normalization/english_normalization.py b/tts/text/pipeline_components/normalization/english_normalization.py new file mode 100644 index 0000000..0c97724 --- /dev/null +++ b/tts/text/pipeline_components/normalization/english_normalization.py @@ -0,0 +1,17 @@ +from tts.text.pipeline_components.normalization.base_normalization import BaseNormalization +from tts.text.pipeline_components.normalization.common_cleaners import collapse_whitespace +from tts.text.pipeline_components.normalization.common_cleaners import convert_to_ascii +from tts.text.pipeline_components.normalization.common_cleaners import lowercase + +class EnglishNormalization(BaseNormalization): + + def normalize(self, raw_input_text: str) -> str: + text = convert_to_ascii(raw_input_text) + text = lowercase(text) + + # TODO: Expand abbreviations + # TODO: Expand numbers + + text = collapse_whitespace(text) + + return text diff --git a/tts/text/normalization/spanish_normalization.py b/tts/text/pipeline_components/normalization/spanish_normalization.py similarity index 58% rename from tts/text/normalization/spanish_normalization.py rename to tts/text/pipeline_components/normalization/spanish_normalization.py index 7475237..56d6712 100644 --- a/tts/text/normalization/spanish_normalization.py +++ b/tts/text/pipeline_components/normalization/spanish_normalization.py @@ -1,4 +1,4 @@ -from tts.text.normalization.base_normalization import BaseNormalization +from tts.text.pipeline_components.normalization.base_normalization import BaseNormalization class SpanishNormalization(BaseNormalization): diff --git a/tts/text/normalization/test_english_normalization.py b/tts/text/pipeline_components/normalization/test_english_normalization.py similarity index 81% rename from tts/text/normalization/test_english_normalization.py rename to tts/text/pipeline_components/normalization/test_english_normalization.py index 4e19092..ecf0d7a 100644 --- a/tts/text/normalization/test_english_normalization.py +++ b/tts/text/pipeline_components/normalization/test_english_normalization.py @@ -1,4 +1,4 @@ -from tts.text.normalization.english_normalization import EnglishNormalization +from tts.text.pipeline_components.normalization.english_normalization import EnglishNormalization class TestEnglishNormalization: diff --git a/tts/text/phonetization/__init__.py b/tts/text/pipeline_components/phonetization/__init__.py similarity index 100% rename from tts/text/phonetization/__init__.py rename to tts/text/pipeline_components/phonetization/__init__.py diff --git a/tts/text/phonetization/base_phonetization.py b/tts/text/pipeline_components/phonetization/base_phonetization.py similarity index 87% rename from tts/text/phonetization/base_phonetization.py rename to tts/text/pipeline_components/phonetization/base_phonetization.py index 73f4115..d8c2ca5 100644 --- a/tts/text/phonetization/base_phonetization.py +++ b/tts/text/pipeline_components/phonetization/base_phonetization.py @@ -1,7 +1,7 @@ import re from typing import List -from tts.text.phonetization.cmudict import CMUDict +from tts.text.pipeline_components.phonetization.cmudict import CMUDict # Regex taken from open source tacotron2 extensions _arpabet_split_re = re.compile(r'[^\s]*{.*?}[^\s]*|[\w]+|[^\w\s]+') diff --git a/tts/text/phonetization/cmudict.py b/tts/text/pipeline_components/phonetization/cmudict.py similarity index 100% rename from tts/text/phonetization/cmudict.py rename to tts/text/pipeline_components/phonetization/cmudict.py diff --git a/tts/text/phonetization/english_phonetization.py b/tts/text/pipeline_components/phonetization/english_phonetization.py similarity index 79% rename from tts/text/phonetization/english_phonetization.py rename to tts/text/pipeline_components/phonetization/english_phonetization.py index 10dd3f2..d09fa7e 100644 --- a/tts/text/phonetization/english_phonetization.py +++ b/tts/text/pipeline_components/phonetization/english_phonetization.py @@ -1,7 +1,7 @@ from typing import List -from tts.text.input_parsing.parser import ArpabetSequence, GraphemeSequence, ParsedSequence -from tts.text.phonetization.base_phonetization import BasePhonetization -from tts.text.phonetization.cmudict import CMUDict +from tts.text.pipeline_components.input_parsing.parser import ArpabetSequence, GraphemeSequence, ParsedSequence +from tts.text.pipeline_components.phonetization.base_phonetization import BasePhonetization +from tts.text.pipeline_components.phonetization.cmudict import CMUDict class EnglishPhonetization(BasePhonetization): diff --git a/tts/text/phonetization/test_cmudict.py b/tts/text/pipeline_components/phonetization/test_cmudict.py similarity index 72% rename from tts/text/phonetization/test_cmudict.py rename to tts/text/pipeline_components/phonetization/test_cmudict.py index 10228bb..bd7963d 100644 --- a/tts/text/phonetization/test_cmudict.py +++ b/tts/text/pipeline_components/phonetization/test_cmudict.py @@ -1,5 +1,5 @@ -from tts.text.phonetization.cmudict import CMUDict -from tts.text.phonetization.cmudict import DEFAULT_CMUDICT_PATH +from tts.text.pipeline_components.phonetization.cmudict import CMUDict +from tts.text.pipeline_components.phonetization.cmudict import DEFAULT_CMUDICT_PATH class TestCmuDict: diff --git a/tts/text/phonetization/test_english_phonetization.py b/tts/text/pipeline_components/phonetization/test_english_phonetization.py similarity index 88% rename from tts/text/phonetization/test_english_phonetization.py rename to tts/text/pipeline_components/phonetization/test_english_phonetization.py index 5b4b4b5..8918aaf 100644 --- a/tts/text/phonetization/test_english_phonetization.py +++ b/tts/text/pipeline_components/phonetization/test_english_phonetization.py @@ -1,7 +1,7 @@ -from tts.text.phonetization.cmudict import DEFAULT_CMUDICT_PATH, CMUDict -from tts.text.phonetization.english_phonetization import EnglishPhonetization -from tts.text.input_parsing.parser import ArpabetSequence -from tts.text.input_parsing.parser import GraphemeSequence +from tts.text.pipeline_components.phonetization.cmudict import DEFAULT_CMUDICT_PATH, CMUDict +from tts.text.pipeline_components.phonetization.english_phonetization import EnglishPhonetization +from tts.text.pipeline_components.input_parsing.parser import ArpabetSequence +from tts.text.pipeline_components.input_parsing.parser import GraphemeSequence class TestEnglishPhonetization: diff --git a/tts/text/sequencing/__init__.py b/tts/text/pipeline_components/sequencing/__init__.py similarity index 100% rename from tts/text/sequencing/__init__.py rename to tts/text/pipeline_components/sequencing/__init__.py diff --git a/tts/text/sequencing/english_symbols_v1.py b/tts/text/pipeline_components/sequencing/english_symbols_v1.py similarity index 96% rename from tts/text/sequencing/english_symbols_v1.py rename to tts/text/pipeline_components/sequencing/english_symbols_v1.py index 3ff0d3d..c947a19 100644 --- a/tts/text/sequencing/english_symbols_v1.py +++ b/tts/text/pipeline_components/sequencing/english_symbols_v1.py @@ -13,7 +13,7 @@ Original documentation: _characters. See TRAINING_DATA.md for details. """ -from tts.text.phonetization import cmudict +from tts.text.pipeline_components.phonetization import cmudict from typing import List # NB: DO NOT CHANGE THE ORDER OF SYMBOLS! diff --git a/tts/text/sequencing/test_english_symbols_v1.py b/tts/text/pipeline_components/sequencing/test_english_symbols_v1.py similarity index 90% rename from tts/text/sequencing/test_english_symbols_v1.py rename to tts/text/pipeline_components/sequencing/test_english_symbols_v1.py index 6c9e71c..b592432 100644 --- a/tts/text/sequencing/test_english_symbols_v1.py +++ b/tts/text/pipeline_components/sequencing/test_english_symbols_v1.py @@ -1,4 +1,4 @@ -from tts.text.sequencing.english_symbols_v1 import arpabet_to_sequence, symbols_to_sequence +from tts.text.pipeline_components.sequencing.english_symbols_v1 import arpabet_to_sequence, symbols_to_sequence class TestEnglishSymbols: