Brummel
b2b1368902
Introduce a new "reset" action for bulk operations and a dedicated endpoint for individual case resets. These features allow administrators to revert a case to its raw audio state by deleting derived artifacts like transcripts, analysis input, and documents. The functionality also includes renaming `.m4a.failed` files back to `.m4a` to re-trigger transcription. This commit also: - Adds a `reset_case_artefacts` helper function to `case_actions.rs`. - Implements the `handle_reset_case` endpoint. - Adds the "Reset" button to the UI for administrators. - Includes unit and integration tests to verify the new functionality and access control.
doctate-whisper
Thin FastAPI wrapper around faster-whisper.
Why a custom service?
Existing Whisper HTTP wrappers (ahmetoner/whisper-asr-webservice,
speaches, linuxserver/faster-whisper, hwdsl2/docker-whisper) hardcode
the faster-whisper defaults. Those defaults include:
condition_on_previous_text=True— classic cascading-hallucination source- temperature fallback cascade
[0.0, 0.2, …, 1.0]— when confidence drops, Whisper gets "creative"
On real cardiology dictations we saw 2/13 runs with catastrophic hallucinations (Russian/Chinese fragments, repetition loops) using ahmetoner's defaults. None of the available wrappers expose these parameters as request fields or env vars.
This service fixes them to safe values that never change.
Anti-hallucination params (fixed)
| Param | Value | Reason |
|---|---|---|
temperature |
0.0 |
No fallback cascade; deterministic output |
condition_on_previous_text |
False |
Prevents cascading hallucinations |
vad_filter |
True |
Silence → no transcription (not garbage) |
vad_parameters.min_silence_duration_ms |
500 |
VAD sensitivity |
no_speech_threshold |
0.6 |
Drop silent segments |
log_prob_threshold |
-1.0 |
Drop low-confidence segments |
compression_ratio_threshold |
2.4 |
Anti-repetition |
beam_size / best_of |
5 / 5 |
Default quality beam search |
Endpoints
POST /asr— multipartaudio_file, queryoutput=txt|json,language=de, forminitial_promptGET /health— liveness:{"status":"ok","model":...,"device":...}GET /info— debug: model, device, compute_type, offline flag
Compatible with ahmetoner's /asr interface so the Axum server needs no change.
Env vars
| Var | Default | Purpose |
|---|---|---|
WHISPER_MODEL |
large-v3-turbo |
Any faster-whisper model name |
WHISPER_COMPUTE_TYPE |
float16 |
float16, int8_float16, int8 (CPU) |
WHISPER_DEVICE |
cuda |
cuda or cpu |
WHISPER_MODELS_DIR |
/models |
Cache location |
WHISPER_OFFLINE |
0 |
1 → no HuggingFace download |
Model choice
Default is large-v3-turbo: same 32-layer encoder as large-v3,
only decoder distilled to 4 layers. OpenAI benchmarks: equal quality on
European languages. ~1.6 GB VRAM vs. ~3 GB — leaves room for Ollama
(gemma4, 9 GB) in a 12 GB card.
Switch to large-v3 at any time via WHISPER_MODEL=large-v3 + container restart.
Build
docker build -t doctate-whisper ./whisper
Takes 2–5 minutes. First run of the container pulls ~1.6 GB from
HuggingFace into /models (volume-persisted).
Deploy to minerva via Dockge
Transport the image without a registry:
docker save doctate-whisper | ssh minerva 'docker load'
Then add the compose file in Dockge and start.
Integration with Axum server
In server/.env:
WHISPER_URL=http://minerva.lan:9001
No other change. The Axum server speaks ahmetoner's interface; we mirror it.
Offline mode
# Pre-pull on a host with internet:
docker run --rm -v /opt/stacks/doctate-whisper/models:/models doctate-whisper \
python3 -c "from faster_whisper import WhisperModel; WhisperModel('large-v3-turbo', download_root='/models')"
# Then set in docker-compose.yml:
environment:
- WHISPER_OFFLINE=1