Generating bidirectional Anki decks with Python from a CSV file

readme.md

1. One-time setup local dev env:

   git clone [email protected]:c683d476bb75cf528da2114d0ffe2bda.git csv-to-anki-deck
   cd csv-to-anki-deck
   python3 -m venv venv
   source venv/bin/activate
   python3 -m pip install genanki gtts
   mkdir audio
   echo -e ".gitignore\nvenv\naudio\ndeck.csv\ndeck.apkg" > .gitignore

2. Activate venv (for every new terminal window/session or OS reboot):

   source venv/bin/activate
   

3. Put a CSV file (e.g. deck.csv) in the root folder of this repo:

   front_word,front_sentence,back_word,back_sentence
   Hallo,"Hallo, ich bin Nora.",Hallo,"Hallo, ik ben Nora."
   wer,Wer bist du?,wie,Wie ben je?
   

4. Generate a unique ID's for your deck:

   python3 -c "import random; print(random.randrange(1 << 30, 1 << 31))"

5. Replace the 1337 deck ID with your generated ID:

   < genanki.Deck(1337, ...)
   ---
   > genanki.Deck(1749144902, ...)

6. Run script, specify CSV file, this creates a .apkg file:

   python3 app.py deck.csv

app.py

import csv
import genanki
from gtts import gTTS
import unicodedata
import re
import os
import sys
from pathlib import Path

if len(sys.argv) < 2:
  print('Usage: python3 app.py <filename>')
  sys.exit(1)

csv_filename = sys.argv[1]
csv_basename = Path(csv_filename).stem

media_files = []

def safe_filename(text):
  text = unicodedata.normalize('NFKD', text).encode('ascii', 'ignore').decode('ascii')
  text = re.sub(r'[^\w\s-]', '', text).strip().lower()
  return re.sub(r'[-\s]+', '-', text)

model = genanki.Model(
  1929277496, # Unique ID for each model
  'Bidirectional word+sentence',
  fields = [
    { 'name': 'front_word' },
    { 'name': 'front_sentence' },
    { 'name': 'back_word' },
    { 'name': 'back_sentence' },
    { 'name': 'front_audio' },
  ],
  templates = [
    {
      'name': 'Front to Back',
      'qfmt': '<strong>{{front_word}}</strong><br><br>{{front_sentence}}<br><br{{front_audio}}',
      'afmt': '<strong>{{front_word}}</strong><br><br>{{front_sentence}}<br><br><hr id="answer"><br><strong>{{back_word}}</strong><br><br>{{back_sentence}}',
    },
    {
      'name': 'Back to Front',
      'qfmt': '<strong>{{back_word}}</strong><br><br>{{back_sentence}}',
      'afmt': '<strong>{{back_word}}</strong><br><br>{{back_sentence}}<br><br><hr id="answer"><br><strong>{{front_word}}</strong><br><br>{{front_sentence}}<br><br{{front_audio}}',
    },
  ]
)

deck = genanki.Deck(
  1337, # Unique ID for each deck
  'Deck' # Name for your deck
)

with open(csv_filename) as csvfile:
  reader = csv.DictReader(csvfile)
  for row in reader:

    fields = [
      row['front_word'],
      row['front_sentence'],
      row['back_word'],
      row['back_sentence'],
    ]

    if row['front_sentence']:
      audio_filename = safe_filename(row['front_sentence']) + '.mp3'

      if not os.path.exists('./audio/' + audio_filename):
        tts = gTTS(text = row['front_sentence'], lang = 'de')
        tts.save('./audio/' + audio_filename)
        print('Generated audio: ' + audio_filename)

      media_files.append('./audio/' + audio_filename)
      fields.append(f'[sound:{audio_filename}]')
    else:
      fields.append('')

    deck.add_note(genanki.Note(
      model = model,
      fields = fields
    ))

genanki.Package(deck, media_files = media_files).write_to_file('./' + csv_basename + '.apkg')