thomas.kopp 658f9be47f fix: punctuate raw transcript, strip JSON code fences, filter null speaker names ...

- llm: punctuate() adds punctuation/capitalisation without changing words
- llm: _strip_code_fences() handles markdown-wrapped JSON from gemma3
- llm: filter string 'null' from identify_speakers result
- pipeline: punctuate raw_text in parallel with refine for solo recordings

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

2026-04-02 12:23:25 +02:00

api

fix: punctuate raw transcript, strip JSON code fences, filter null speaker names

2026-04-02 12:23:25 +02:00

docs

feat: AI-generated title+tldr, subfolder structure, backlinks in transkript/zusammenfassung

2026-04-02 12:07:54 +02:00

frontend

feat: tab navigation in modal (Index/Transkript/Zusammenfassung)

2026-04-02 12:10:50 +02:00

tests

feat: AI-generated title+tldr, subfolder structure, backlinks in transkript/zusammenfassung

2026-04-02 12:07:54 +02:00

.gitignore

chore: project scaffold

2026-04-01 02:12:41 +02:00

alignment.py

feat: align_segments() — map Whisper timestamps to pyannote speakers

2026-04-02 01:00:58 +02:00

audio.py

fix: record at 48000 Hz — PipeWire virtual sinks reject 16 kHz resampling

2026-04-02 09:14:34 +02:00

auth.py

feat: multi-user auth — per-user spaces, pbkdf2 passwords, session tokens, login page

2026-04-01 08:39:16 +02:00

CLAUDE.md

chore: project scaffold

2026-04-01 02:12:41 +02:00

config.py

feat: copy transcript to Obsidian vault on open

2026-04-02 11:00:55 +02:00

diarization.py

fix: handle pyannote 4.x DiarizeOutput wrapper in diarize()

2026-04-02 11:34:02 +02:00

install.sh

fix: use ip route instead of hostname -I for Arch compatibility

2026-04-01 12:23:41 +02:00

llm.py

fix: punctuate raw transcript, strip JSON code fences, filter null speaker names

2026-04-02 12:23:25 +02:00

main.py

fix: use sounddevice names for audio device list and combined source

2026-04-02 07:51:42 +02:00

output.py

feat: tab navigation in modal (Index/Transkript/Zusammenfassung)

2026-04-02 12:10:50 +02:00

pytest.ini

chore: project scaffold

2026-04-01 02:12:41 +02:00

requirements.txt

feat: Diarizer class wrapping pyannote/speaker-diarization-3.1

2026-04-02 00:59:50 +02:00

transcription.py

feat: add whisper.cpp ROCm backend support for AMD GPU acceleration

2026-04-02 01:33:32 +02:00

S

Description

Lokales KI-Transkriptionstool — Whisper + Ollama + Browser-UI

196 KiB

Languages

Python 63.7%

HTML 21.7%

JavaScript 12.3%

Shell 2.3%