Publish artifacts for audio-event-triage-20260902

README.md

@@ -0,0 +1,59 @@
+---
+license: cc-by-4.0
+language:
+- en
+pretty_name: Audio Event Triage Baseline Synthetic Evaluation Set
+size_categories:
+- n<1K
+task_categories:
+- audio-classification
+tags:
+- synthetic
+- audio-ml
+- evaluation
+- audio-classification
+- automatic-speech-recognition
+- feature-extraction
+- audio-to-audio
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train.jsonl
+  - split: test
+    path: data/test.jsonl
+---
+
+# Audio Event Triage Baseline Synthetic Dataset
+
+## Summary
+
+This dataset contains 14 training examples and 4
+held-out examples for **Operations teams need an explainable starting point for classifying alarms, machinery noise, and speech-like events.**
+
+Every record is synthetic and includes:
+
+- `input`: query, event, or feature description
+- `label`: expected class, route, relation, or evidence category
+- `context`: synthetic supporting context
+- `source`: fictional source identifier
+- `variant`: generation pattern
+- `synthetic`: always `true`
+
+## Uses
+
+- Reproducible unit and integration tests
+- Baseline model training
+- Evaluation harness development
+- Schema and architecture demonstrations
+
+## Limitations
+
+The included records are synthetic feature vectors and do not replace evaluation on licensed real audio.
+
+This dataset does not represent real users, patients, customers, production
+traffic, or licensed media. It must not be presented as real-world evidence.
+
+## Related Model
+
+[RKB109/audio-event-triage-20260902-model](https://huggingface.co/RKB109/audio-event-triage-20260902-model)

data/test.jsonl

@@ -0,0 +1,4 @@
+{"id":"audio-event-triage-2-1","input":"low frequency continuous vibration high zero crossing stability","label":"machinery","context":"Synthetic rotating equipment feature summary.","source":"audio-02","variant":"direct","synthetic":true}
+{"id":"audio-event-triage-3-2","input":"In an operations review, variable pitch moderate energy speech cadence","label":"speech","context":"Synthetic spoken segment feature summary.","source":"audio-03","variant":"operations","synthetic":true}
+{"id":"audio-event-triage-4-3","input":"For an evaluation case, repeating pulse bright spectrum urgent tone","label":"alarm","context":"Synthetic alert feature summary.","source":"audio-04","variant":"evaluation","synthetic":true}
+{"id":"audio-event-triage-6-1","input":"syllabic rhythm changing pitch voice activity","label":"speech","context":"Synthetic voice feature summary.","source":"audio-06","variant":"direct","synthetic":true}

data/train.jsonl

@@ -0,0 +1,14 @@
+{"id":"audio-event-triage-1-1","input":"high energy repeating tone peak frequency 2100","label":"alarm","context":"Synthetic alarm feature summary.","source":"audio-01","variant":"direct","synthetic":true}
+{"id":"audio-event-triage-1-2","input":"In an operations review, high energy repeating tone peak frequency 2100","label":"alarm","context":"Synthetic alarm feature summary.","source":"audio-01","variant":"operations","synthetic":true}
+{"id":"audio-event-triage-1-3","input":"For an evaluation case, high energy repeating tone peak frequency 2100","label":"alarm","context":"Synthetic alarm feature summary.","source":"audio-01","variant":"evaluation","synthetic":true}
+{"id":"audio-event-triage-2-2","input":"In an operations review, low frequency continuous vibration high zero crossing stability","label":"machinery","context":"Synthetic rotating equipment feature summary.","source":"audio-02","variant":"operations","synthetic":true}
+{"id":"audio-event-triage-2-3","input":"For an evaluation case, low frequency continuous vibration high zero crossing stability","label":"machinery","context":"Synthetic rotating equipment feature summary.","source":"audio-02","variant":"evaluation","synthetic":true}
+{"id":"audio-event-triage-3-1","input":"variable pitch moderate energy speech cadence","label":"speech","context":"Synthetic spoken segment feature summary.","source":"audio-03","variant":"direct","synthetic":true}
+{"id":"audio-event-triage-3-3","input":"For an evaluation case, variable pitch moderate energy speech cadence","label":"speech","context":"Synthetic spoken segment feature summary.","source":"audio-03","variant":"evaluation","synthetic":true}
+{"id":"audio-event-triage-4-1","input":"repeating pulse bright spectrum urgent tone","label":"alarm","context":"Synthetic alert feature summary.","source":"audio-04","variant":"direct","synthetic":true}
+{"id":"audio-event-triage-4-2","input":"In an operations review, repeating pulse bright spectrum urgent tone","label":"alarm","context":"Synthetic alert feature summary.","source":"audio-04","variant":"operations","synthetic":true}
+{"id":"audio-event-triage-5-1","input":"steady hum harmonic vibration low pitch","label":"machinery","context":"Synthetic motor feature summary.","source":"audio-05","variant":"direct","synthetic":true}
+{"id":"audio-event-triage-5-2","input":"In an operations review, steady hum harmonic vibration low pitch","label":"machinery","context":"Synthetic motor feature summary.","source":"audio-05","variant":"operations","synthetic":true}
+{"id":"audio-event-triage-5-3","input":"For an evaluation case, steady hum harmonic vibration low pitch","label":"machinery","context":"Synthetic motor feature summary.","source":"audio-05","variant":"evaluation","synthetic":true}
+{"id":"audio-event-triage-6-2","input":"In an operations review, syllabic rhythm changing pitch voice activity","label":"speech","context":"Synthetic voice feature summary.","source":"audio-06","variant":"operations","synthetic":true}
+{"id":"audio-event-triage-6-3","input":"For an evaluation case, syllabic rhythm changing pitch voice activity","label":"speech","context":"Synthetic voice feature summary.","source":"audio-06","variant":"evaluation","synthetic":true}

project.json

@@ -0,0 +1,52 @@
+{
+  "name": "Audio Event Triage Baseline",
+  "problem": "Operations teams need an explainable starting point for classifying alarms, machinery noise, and speech-like events.",
+  "domain": "audio-ml",
+  "architecture": "classifier",
+  "hugging_face_tasks": [
+    "audio-classification",
+    "automatic-speech-recognition",
+    "feature-extraction",
+    "audio-to-audio"
+  ],
+  "recommended_stack": [
+    "FastAPI for audio metadata and prediction APIs",
+    "Transformers with Wav2Vec2 or audio spectrogram models",
+    "librosa or torchaudio for feature extraction",
+    "MLflow for experiment and model registry",
+    "Object storage for licensed audio",
+    "OpenTelemetry for inference latency and error traces"
+  ],
+  "real_world_data_sources": [
+    {
+      "name": "Freesound API",
+      "url": "https://freesound.org/apiv2/search/text/?query=alarm&page_size=5",
+      "purpose": "Licensed public audio search when an API key is configured"
+    },
+    {
+      "name": "Hugging Face Datasets API",
+      "url": "https://huggingface.co/api/datasets?search=audio-classification&limit=5",
+      "purpose": "Discover public audio datasets and metadata"
+    }
+  ],
+  "job_description_skills": [
+    "Audio feature pipelines and model fine-tuning",
+    "Class imbalance and macro-metric evaluation",
+    "Streaming inference and confidence calibration",
+    "Dataset licensing and provenance controls",
+    "Model monitoring and human review workflows"
+  ],
+  "impact_targets": [
+    "Reach macro recall >= 0.85 on a licensed evaluation set",
+    "Review every prediction below the confidence threshold",
+    "Process one minute of audio in under five seconds",
+    "Track class-level drift and false-negative rates"
+  ],
+  "baseline_evaluation": {
+    "test_examples": 4,
+    "accuracy": 1,
+    "synthetic_evaluation": true
+  },
+  "estimated_delivery": "8-12 weeks for one engineer",
+  "generated_baseline_is_production_ready": false
+}

sources.json

@@ -0,0 +1,12 @@
+[
+  {
+    "name": "Freesound API",
+    "url": "https://freesound.org/apiv2/search/text/?query=alarm&page_size=5",
+    "purpose": "Licensed public audio search when an API key is configured"
+  },
+  {
+    "name": "Hugging Face Datasets API",
+    "url": "https://huggingface.co/api/datasets?search=audio-classification&limit=5",
+    "purpose": "Discover public audio datasets and metadata"
+  }
+]