88 lines
3.8 KiB
Markdown
88 lines
3.8 KiB
Markdown
# German Learning — Co-learning Context Repo
|
|
|
|
This repo tracks what I've actually covered across German-learning materials
|
|
(course books, workbooks, audio transcripts, and answer keys), lesson by
|
|
lesson, so that Claude (or any LLM) has exact context on my vocabulary and
|
|
grammar level when we study together. *Menschen A1.1* is the first source;
|
|
later books can be added alongside it.
|
|
|
|
## How to use this with Claude
|
|
|
|
At the start of a study session, point Claude at the relevant lesson file(s)
|
|
(e.g. `A1.1/lektion-05.md`), or the whole `A1.1/INDEX.md` for a running
|
|
summary. Ask Claude to only use vocabulary/grammar that appears in those
|
|
files (or earlier lessons) when creating exercises, example sentences, or
|
|
corrections — this keeps practice within what I've actually learned.
|
|
|
|
## Structure
|
|
|
|
```
|
|
A1.1/
|
|
lektion-XX.md ← everything for one lesson: vocab, grammar, communication
|
|
patterns, course book + workbook exercise summaries,
|
|
official audio transcripts, answer key highlights
|
|
_template.md ← template used to build each lektion-XX.md
|
|
INDEX.md ← one-line summary of every lesson so far
|
|
A1.2/ ← added once I move on to the next book
|
|
...
|
|
scripts/
|
|
transcribe.py ← local ASR fallback (faster-whisper) for audio with no
|
|
official transcript
|
|
```
|
|
|
|
Each lesson is one combined file rather than separate course-book/workbook
|
|
files, because *Menschen*'s workbook exercises constantly reference course
|
|
book sections (e.g. "KB 2b") and share the same grammar — splitting them
|
|
apart would just duplicate content.
|
|
|
|
## Adding a new lesson
|
|
|
|
1. Get the course book pages, workbook pages, and (if available) the
|
|
official Hueber Transkriptionen PDF pages for that lesson — as PDFs,
|
|
photos, or a folder path.
|
|
2. Claude extracts/summarizes content into `lektion-XX.md` using
|
|
`_template.md`: vocabulary (from the workbook's Lernwortschatz page),
|
|
grammar boxes, communication patterns, a brief summary of what each
|
|
exercise practices (not full exercise text), audio transcripts, and
|
|
answer key highlights.
|
|
3. Review and correct anything Claude got wrong or over-summarized.
|
|
4. Update `INDEX.md` with a one-line entry for the lesson.
|
|
5. Commit.
|
|
|
|
## Audio (listening exercises)
|
|
|
|
Hueber releases official **Transkriptionen** PDFs for both the Kursbuch
|
|
(course book) and Arbeitsbuch (workbook) — these are the primary source for
|
|
audio content, not the *Lösungsschlüssel* (answer key — a different
|
|
document). Raw audio files are **not** stored in this repo (see
|
|
`.gitignore`); only transcripts (text) are, since that's what gives Claude
|
|
context.
|
|
|
|
Primary workflow — official transcripts:
|
|
1. Get the Kursbuch and Arbeitsbuch Transkriptionen PDFs for the relevant
|
|
book (A1.1, A1.2, ...).
|
|
2. Extract the per-track dialogue text into that lesson's `.md` file under
|
|
"Audio / Listening", marked as official-sourced.
|
|
3. No review needed for accuracy — these are publisher-official text, not ASR.
|
|
|
|
Fallback workflow — local ASR (only if no official transcript exists for a
|
|
given track, e.g. supplementary listening material):
|
|
1. Get the audio files locally (publisher CD/app/download), any folder
|
|
outside git.
|
|
2. Run `scripts/transcribe.py` (uses `faster-whisper` on GPU) to transcribe
|
|
them:
|
|
```bash
|
|
pip install faster-whisper
|
|
python scripts/transcribe.py "<path-to-lesson-audio>" "A1.1/transcripts/lektion-XX"
|
|
```
|
|
3. Skim the `.txt` output for ASR mistakes (names, numbers, fast speech) and
|
|
fix them.
|
|
4. Reference/paste the transcript into the lesson's `.md` file under "Audio
|
|
/ Listening", flagged as ASR-generated — transcripts are small text
|
|
files so committing them is fine.
|
|
|
|
## Adding a new book (e.g. A1.2)
|
|
|
|
Copy the `A1.1/` folder structure (minus content) into a new `A1.2/` folder
|
|
and repeat the process.
|