add dictation model

src/anki_hsk_creator/anki_generation.py

@@ -64,6 +64,23 @@ PHRASE_MODEL = Model(
     css=CSS,
 )
 
+DICTATION_MODEL = Model(
+    3187277536,
+    "Phrase Model",
+    fields=[
+        {"name": "Translated"},
+        {"name": "Phrase"},
+        {"name": "Audio"},
+    ],
+    templates=[
+        {
+            "name": "Card 1",
+            "qfmt": "{{Audio}}<br>{{type:Phrase}}",
+            "afmt": '{{FrontSide}}<hr id="answer">{{Phrase}}<br>{{Translated}}',
+        },
+    ],
+    css=CSS,
+)
 
 HSK_MODEL = Model(
     1708536519,
@@ -122,16 +139,52 @@ HSK_MODEL = Model(
 # Proccess
 
 
+def output_anki_dictation(
+    process_file: ProcessFile, results: list[DictionaryResult]
+) -> Path:
+    """Creates an anki file for dictation result"""
+    final_file = process_file.output_name.with_suffix(".apkg")
+    deck_name = "::".join(
+        process_file.input_file.parts[:-1] + (process_file.output_name.stem,)
+    )
+    deck = Deck(
+        random.randrange(1 << 30, 1 << 31),
+        deck_name,
+        f"Deck for {final_file.name}, "
+        "created in https://www.wolfang.info.ve/hskankicreator/",
+    )
+    package = Package(deck)
+    audios = []
+    for result in results:
+        note = Note(
+            model=DICTATION_MODEL,
+            fields=[
+                result.translated,
+                result.line,
+                f"[sound:{result.audio_path.name}]",
+            ],
+        )
+        deck.add_note(note)
+        audios.append(result.audio_path)
+    package.media_files = audios
+    package.write_to_file(final_file)
+    return final_file
+
+
 def output_anki_dictionary(
     process_file: ProcessFile, results: list[DictionaryResult]
 ) -> Path:
     """Creates an anki file from a dictionary results"""
     final_file = process_file.output_name.with_suffix(".apkg")
-    print()
     deck_name = "::".join(
         process_file.input_file.parts[:-1] + (process_file.output_name.stem,)
     )
-    deck = Deck(random.randrange(1 << 30, 1 << 31), deck_name)
+    deck = Deck(
+        random.randrange(1 << 30, 1 << 31),
+        deck_name,
+        f"Deck for {final_file.name}, "
+        "created in https://www.wolfang.info.ve/hskankicreator/",
+    )
     package = Package(deck)
     audios = []
     for result in results:
@@ -161,7 +214,12 @@ def output_anki_phrase(
     deck_name = "::".join(
         process_file.input_file.parts[:-1] + (process_file.output_name.stem,)
     )
-    deck = Deck(random.randrange(1 << 30, 1 << 31), deck_name)
+    deck = Deck(
+        random.randrange(1 << 30, 1 << 31),
+        deck_name,
+        f"Deck for {final_file.name}, "
+        "created in https://www.wolfang.info.ve/hskankicreator/",
+    )
     package = Package(deck)
     audios = []
     for result in results:

src/anki_hsk_creator/api.py

@@ -8,9 +8,26 @@ from pathlib import Path
 
 # Local
 from . import DATA_FOLDER
-from .anki_generation import output_anki_dictionary, output_anki_phrase
-from .constants import DICT_TYPE, INPUT, LANGUAGES, OUTPUT, PHRASES_TYPE, RESOURCES
-from .proccessor import dictionary_pre_process, dictionary_process, translator_process
+from .anki_generation import (
+    output_anki_dictation,
+    output_anki_dictionary,
+    output_anki_phrase,
+)
+from .constants import (
+    DICTATION_TYPE,
+    DICT_TYPE,
+    INPUT,
+    LANGUAGES,
+    OUTPUT,
+    PHRASES_TYPE,
+    RESOURCES,
+)
+from .proccessor import (
+    dictation_process,
+    dictionary_pre_process,
+    dictionary_process,
+    translator_process,
+)
 from .utility import CCCEDICT, TRANS, TTS, ProcessFile
 
 # interface
@@ -74,7 +91,7 @@ def create_folder(file_path: Path) -> ProcessFile:
 
 
 def delete_folder(file_path: Path):
-    """delete an empty folder in file_path"""
+    """Delete an empty folder in file_path"""
     input_folder = INPUT / file_path
     if input_folder.exists():
         if any(Path("some/path/here").iterdir()):
@@ -136,10 +153,12 @@ def create_input_file(
     it is created and the file placed inside.
     returns the relative path for future processing
 
-    valid file_types: ".phrases", ".dictionary"
+    valid file_types: ".phrases", ".dictionary" ".dictation"
     """
-    if file_type not in (PHRASES_TYPE, DICT_TYPE):
-        raise ValueError(f"file_type {file_type} not in {(PHRASES_TYPE, DICT_TYPE)}")
+    if file_type not in (PHRASES_TYPE, DICT_TYPE, DICTATION_TYPE):
+        raise ValueError(
+            f"file_type {file_type} not in {(PHRASES_TYPE, DICT_TYPE, DICTATION_TYPE)}"
+        )
     filename = f"{name}{file_type}.txt"
     relative = sub_folder / filename
     # write file
@@ -152,6 +171,7 @@ def create_input_file(
 
 
 def write_input_file(process_file: ProcessFile, text: str):
+    """Write an input file"""
     with process_file.absolute_input_file.open(
         "w", encoding="utf8", newline="\n"
     ) as file:
@@ -162,6 +182,7 @@ def write_input_file(process_file: ProcessFile, text: str):
 
 
 def write_resource_file(process_file: ProcessFile, language_id: str, text: str):
+    """Write a resource file"""
     process_file.language_id = language_id
     with process_file.dictionary_resource_file.open(
         "w", encoding="utf8", newline="\n"
@@ -184,13 +205,28 @@ def pre_process_a_dictionary_file(process_file: ProcessFile, language_id: str):
 
 
 def process_a_dictionary_file(process_file: ProcessFile, language_id: str) -> Path:
+    """Process a dictionary file"""
     TTS.create_tts()
     process_file.language_id = language_id
     results = dictionary_process(process_file)
     return output_anki_dictionary(process_file, results)
 
 
+def process_a_dictation_file(process_file: ProcessFile, language_id: str) -> Path:
+    """Process a dictation file"""
+    TTS.create_tts()
+    TRANS.create_translator(LANGUAGES.CN, language_id)
+    process_file.language_id = language_id
+    with process_file.absolute_input_file.open(
+        "r", encoding="utf8", newline="\n"
+    ) as file:
+        text_lines = [line.strip() for line in file.read().split("。")]
+    results = dictation_process(text_lines, process_file)
+    return output_anki_dictation(process_file, results)
+
+
 def process_a_phrases_file(process_file: ProcessFile, language_id: str) -> Path:
+    """Process a phrases file"""
     process_file.language_id = language_id
     TTS.create_tts()
     TRANS.create_translator(LANGUAGES.CN, language_id)

src/anki_hsk_creator/constants.py

@@ -20,6 +20,7 @@ RESOURCES.mkdir(exist_ok=True, parents=True)
 # File Types
 PHRASES_TYPE = ".phrases"
 DICT_TYPE = ".dictionary"
+DICTATION_TYPE = ".dictation"
 
 
 class LANGUAGES:

src/anki_hsk_creator/proccessor.py

@@ -19,6 +19,26 @@ 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):
+        line = line.strip()
+        audio_path = process_file.resources / f"N{n:03n}.wav"
+        if not audio_path.exists():
+            audio = TTS.MODEL.generate(f"{line}。", language_id=LANGUAGES.CN)
+            torchaudio.save(audio_path, audio, TTS.MODEL.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]: