- Conversie paralelă cu ProcessPoolExecutor (4 workers default)
- Tabele: font 8.5pt, padding compact, word-break pentru text lung
- code-friendly extra previne pierderea _ din nume de fișiere
- find_files caută toate *.md din summaries/, nu doar MODUL*
- .gitignore: adaugă .claude/ (local state)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Referințele inventate (ex: "MODUL4 Cercetasi Ex.3") au fost înlocuite
cu căi reale către fișierele sursă și secțiunile exacte
(ex: MODUL4_STRATEGII_APLICARI.md → "Pentru cercetași adulți" → "Exercitiu 3").
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add 41 transcript files and 31 summary files to version control.
Previously excluded via .gitignore — now tracked for backup and portability.
SRT subtitle files remain excluded (derivable from transcripts).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Adaugă al 4-lea nivel de sumarizare: caiete de facilitator cu corelații
cross-disciplinare (IFS, CNV, attachment theory, Brené Brown, neuroștiință),
exerciții cu script complet, și ghiduri de aplicare cu secvențe de cadrare
pentru 4 contexte (grup NLP, cercetași, CNV, self-coaching).
Fișiere noi (în summaries/, ne-tracked de .gitignore):
- 5 caiete: MODUL{1-5}_{CONCEPT}.md
- 5 aplicări: MODUL{1-5}_{CONCEPT}_APLICARI.md
- 5 cross-modul: INDEX_EXERCITII, HARTA_CONEXIUNI, GLOSAR_CREDINTE,
GHID_FACILITARE, METAFORE_POVESTI
Tracked:
- PROMPT_EXPERIENTIAL.md: prompt reutilizabil pentru M6
- TODOS.md: instrucțiuni complete pentru M6 (run.bat 6 + sumarizări + caiet)
- md_to_pdf.py: convertor markdown→PDF pentru summaries
- PROCES_SUMARIZARE.md: documentație proces sumarizare standard
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- transcribe.py: add --max-context 0, --entropy-thold 2.4, --max-len 60,
--suppress-nst, --no-fallback to whisper.cpp to prevent hallucination loops
- transcribe.py: remove interactive quality gate (runs unattended now)
- run.bat: remove pause prompts for unattended operation
- retranscribe_tail.py: new script that detects hallucination bursts in SRT
files, extracts and re-transcribes only the affected audio segments, then
splices the result back together. Drops segments that re-hallucinate
(silence/music). Backs up originals to transcripts/backup/.
- fix_hallucinations.bat: Windows wrapper for retranscribe_tail.py
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add manifest.json to .gitignore so it doesn't conflict between machines.
Also add CLAUDE.md for Claude Code guidance.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Parametrul din run.bat (ex: 4-5) era transmis doar la transcribe.py.
Acum download.py primeste acelasi filtru si descarca doar modulele
specificate. Sintaxa acceptata: '4-5', '1,3', '1-3,5'.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Store stub-ul pentru python.exe termina procesul batch cand e apelat
direct. Fix: foloseste 'py' launcher (safe) sau detecteaza python.exe
real via 'where | findstr /v WindowsApps', fara a executa python
in contextul check-ului. Toate apelurile python -> !PYTHON_CMD!.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- setup_whisper.py: descarcă build CPU din release-urile oficiale,
sare peste Vulkan/CUDA/OpenBLAS
- run.bat: elimină env var GGML_VK_PREFER_HOST_MEMORY și check-ul Vulkan SDK
- transcribe.py: --no-gpu era deja setat
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- run.bat: one-click pipeline (download, convert, transcribe)
- download.py: fetch audio from course platform
- transcribe.py: whisper.cpp batch transcription (CPU, WAV 16kHz)
- MP3->WAV conversion via ffmpeg
- --modules filter for splitting work across machines
- summarize.py: generate summaries from transcripts
- setup_whisper.py: auto-download whisper.cpp, ffmpeg, and model
- Medium model (q5_0) instead of large to avoid VRAM crashes
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
