Back to evidence library

07 / Independent ML systems project · 2026

Tested public repositorydata scienceml systems

Environmental audio clustering as a reproducible system

One canonical feature contract from waveform generation to served cluster assignment.

The original README-only repository is now an executable system. Training and serving share an explicit feature schema, while synthetic signals provide a licensed, deterministic public demo path.

Audio Clustering project visual

Evidence register

170

features

Canonical train/serve schema

4/4

tests

Local and CI suite

ARI

external metric

Computed only when labels exist

01 / Problem

Notebook-only clustering cannot prove that feature extraction, training and inference agree on the same representation.

02 / Approach

A shared extractor validates audio, builds 170 features and feeds both the Kedro training pipeline and bounded inference API.

03 / Outcome

The repository now runs end to end with generated signals, packaging, tests, CI and container documentation.

How the evidence is produced.

Rebuilt the repository into a reproducible package, test suite, CI workflow, container and bounded API.

  1. 01Waveform → validated audio
  2. 02Shared extractor → 170 features
  3. 03Pipeline → clusters and artifact
  4. 04API → bounded assignment

Validation scope

Four tests cover feature shape, deterministic generation and core pipeline contracts. Synthetic labels allow a limited ARI sanity check.

Known limitation

Synthetic separability does not establish usefulness on a real acoustic corpus; stability and representation quality still require external evaluation.

What is inspectable

  • Train and serve import one schema.
  • The public demo needs no opaque dataset.
  • ARI is never fabricated for unlabeled data.

Next proof to add

  1. 01Evaluate on a governed public corpus.
  2. 02Measure stability across seeds and perturbations.
  3. 03Add latency and drift monitoring.

Main stack

PythonKedroAudio featuresClusteringFastAPIDocker