144 lines
5.3 KiB
Python
144 lines
5.3 KiB
Python
"""processor.py"""
|
||
|
||
# Standard Library
|
||
import csv
|
||
|
||
# Pip
|
||
import argostranslate.translate
|
||
import soundfile as sf
|
||
|
||
# Local
|
||
from .constants import LANGUAGES
|
||
from .utility import CCCEDICT, TTS, DictionaryResult, ProcessFile, TranslationResult
|
||
|
||
# Constants
|
||
|
||
FIELDNAMES = ["simplified", "traditional", "pinyin", "meaning"]
|
||
DIALECT = "excel-tab"
|
||
|
||
# Results Classes
|
||
|
||
|
||
def dictation_process(
|
||
text_lines: list[str], process_file: ProcessFile
|
||
) -> list[TranslationResult]:
|
||
"""Process for Dictation translation"""
|
||
results = []
|
||
for n, line in enumerate(text_lines):
|
||
audio_line = line.strip()
|
||
# line = " ".join(line.split())
|
||
# audio_line = " ".join(line)
|
||
# audio_line = audio_line.replace("，", "，。。。]")
|
||
audio_path = process_file.resources / f"N{n:03n}.wav"
|
||
if not audio_path.exists():
|
||
wavs, sr = TTS.generate(f"{audio_line}。")
|
||
sf.write(audio_path, wavs[0], sr)
|
||
translated = argostranslate.translate.translate(
|
||
line, LANGUAGES.CN, process_file.language_id
|
||
)
|
||
results.append(
|
||
TranslationResult(process_file.language_id, translated, line, audio_path)
|
||
)
|
||
return results
|
||
|
||
|
||
def translator_process(
|
||
text_lines: list[str], process_file: ProcessFile
|
||
) -> list[TranslationResult]:
|
||
"""Process for phases or sentence translation"""
|
||
results = []
|
||
for n, line in enumerate(text_lines):
|
||
audio_line = line.strip()
|
||
# line = " ".join(line.split())
|
||
# audio_line = " ".join(line)
|
||
# audio_line = audio_line.replace("，", "，。。。]")
|
||
audio_path = process_file.resources / f"N{n:03n}.wav"
|
||
if not audio_path.exists():
|
||
wavs, sr = TTS.generate(f"{audio_line}。")
|
||
sf.write(audio_path, wavs[0], sr)
|
||
translated = argostranslate.translate.translate(
|
||
line, LANGUAGES.CN, process_file.language_id
|
||
)
|
||
results.append(
|
||
TranslationResult(process_file.language_id, translated, line, audio_path)
|
||
)
|
||
return results
|
||
|
||
|
||
def dictionary_pre_process(words_list: list[str], process_file: ProcessFile):
|
||
"""Pre Process dictionary files into a intermediary resources file"""
|
||
dictionary = CCCEDICT.create_cedict(process_file.language_id)
|
||
with process_file.dictionary_resource_file.open(
|
||
"w", encoding="utf8", newline=""
|
||
) as resource_file:
|
||
tsv_writer = csv.DictWriter(
|
||
resource_file, dialect=DIALECT, fieldnames=FIELDNAMES
|
||
)
|
||
tsv_writer.writeheader()
|
||
for words in words_list:
|
||
word = words.split()[0]
|
||
pinyin = " ".join(words.split()[1:]) if len(words.split()) > 1 else None
|
||
if entries := dictionary.get(word):
|
||
if pinyin is not None:
|
||
entries = list(filter(lambda x: x.pinyin == pinyin, entries))
|
||
if len(entries) > 1:
|
||
print(f"\nWARNING: {word} has multiple meanings:")
|
||
for entry in entries:
|
||
for meaning in entry.meanings:
|
||
tsv_writer.writerow(
|
||
{
|
||
"simplified": entry.simplified,
|
||
"traditional": entry.traditional,
|
||
"pinyin": entry.pinyin,
|
||
"meaning": meaning,
|
||
}
|
||
)
|
||
else:
|
||
print("============================================")
|
||
print(f"===================>ERROR: {word} not found")
|
||
print("============================================")
|
||
tsv_writer.writerow(
|
||
{
|
||
"simplified": word,
|
||
"traditional": None,
|
||
"pinyin": None,
|
||
"meaning": None,
|
||
}
|
||
)
|
||
|
||
|
||
def dictionary_process(process_file: ProcessFile) -> list[DictionaryResult]:
|
||
"""Process a dictionary_resource_file into a final result"""
|
||
results = []
|
||
with process_file.dictionary_resource_file.open(
|
||
"r", encoding="utf8", newline=""
|
||
) as resource_file:
|
||
reader = csv.DictReader(resource_file, dialect=DIALECT)
|
||
for line in reader:
|
||
audio_path = process_file.resources / f"{line['pinyin']}.wav"
|
||
if not audio_path.exists():
|
||
wavs, sr = TTS.generate(f"{line['simplified']}。")
|
||
sf.write(audio_path, wavs[0], sr)
|
||
print(line)
|
||
result = DictionaryResult(
|
||
**line, audio_path=audio_path, language_id=process_file.language_id
|
||
)
|
||
results.append(result)
|
||
return results
|
||
|
||
|
||
# def output_tsv(out_file, results):
|
||
# """writes the output as a tsv file"""
|
||
# final_file = out_file.parent / f"{out_file.stem}.tsv"
|
||
# with final_file.open("w", encoding="utf8", newline="") as csvfile:
|
||
# writer = csv.writer(csvfile, delimiter="\t", quotechar='"')
|
||
# for entry in results:
|
||
# writer.writerow(
|
||
# [
|
||
# "\n ".join(f"{n+1}. {m}" for n, m in enumerate(entry.meanings)),
|
||
# PinyinToneConverter().convert_text(entry.pinyin),
|
||
# entry.simplified,
|
||
# entry.traditional,
|
||
# ]
|
||
# )
|