LangTrainerPy

A more simple version of what my LangTrainer desktop app was aiming to be...

TODO

• Spanish:
  • there are additional tenses that can only be expressed with the subjunctive. This needs to be implemented too, things like "will have eaten", or the variations on the imperative mood for example.
• Possibly consider creating an additional YAML file that defines the word-ending rules rather than hardcoding it in the Python itself. This way it's just less hacky/hardcodey Python functions, and modifying the rules for a language becomes much easier and more modular. You could have files like spanish.yaml, french.yaml, russian.yaml and so on.
  • this could serve as a replacement for the existing function derive_ending()

Default Behaviour

• This is a language practice program, optimised for drilling a given target language. It is not intended for memorisation of vocab, that's a job for Anki. LangTrainer is dead simple, it allows you to enter in (primarily) nouns and verbs for a specific language, and it will then present you these same (primarily) nouns and verbs, asking you to conjugate/decline them accordingly and notifying you when/if you get any wrong. That's it...

Features

• the default behaviour should be that the user can keep practicing over and over until they terminate the program manually, with no card/time limit.

• specifically for German, it'd be useful to also include possessive pronouns since they also decline based on case... Try to add these in as well!

• specifically for Russian, it'd be great to be able to select a specific case (instrumental for example) and then have to decline a bunch of examples of nouns across all the genders to be able to nail my noun declensions.

• should also add all verb tenses too -> past, present, future etc.

• don't forget about participle forms for these languages too!

Training mode

• additional args can be used to narrow down the specific area of study, be it specific verb tenses/conjugations or declining nouns for a certain subset of genders etc, things like that.

Extension Modes

• include a Pimsleur-like mode, that reads out with a TTS engine, the random assortment of sentences (somehow only reading out the valid generated sentences that you'd actually find in the real world), and waiting with a short pause, for the user to repeat out loud. Maybe include the English version, and then make the user say the spoken version out loud before playing it via the TTS engine or something like this. This would force the user to come up with the conjugations and declensions on the spot, which would be invaluable practice.

Roadmap

Currently, I have the remaining tasks to complete before I can use this to drill verbs conjugations:

• Complete the derive_ending() function for all the tenses in Spanish and German. For Spanish, these remaining tenses are: future, pres. perfect (past participle), preterite and imperfect.
• Implement the drill functionality. Iterate over the stored words in the target language and drill the verb conjugations.
  • randomly select an infinitive, randomly select a tense, randomly select a conjugation, present the english verb (to be, will be, was, were etc.) and indicate the subject conjugation. the tense will be indicated by the English word, but we need to indicate the subject conjugation in brackets.
  • might want to mark this combo of infinitive, conjugation, tense as complete so that it doesn't get generated again. If a combination is generated that's already been tested, just continue to the top of the loop and randomly generate again from scratch (the time complexity is negligible)
  • if answered right, then return to top of loop and repeat, if answered wrong then might need to store this combination and assess it later. Might want to indicate how many incorrect cards are in the backlog ready to be re-assessed. This would be handy for the user to know since they might not want to quit the practice session early if they have re-assess cards pending.
• German noun declensions would be handy and the implementation would probably resemble that of the verb conjugations anyway so quite simple.