# adhi-1235/audio-context-layer architecture

> Architecture diagram of the GitHub repository adhi-1235/audio-context-layer, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-02). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/adhi-1235/audio-context-layer
- Repository: https://github.com/adhi-1235/audio-context-layer

## Overview

Audio Context Layer is an exploratory Clotho-AQA research workspace, not a deployed QA service. Its documented workflow inspects annotation splits and audio, groups repeated answers into processed tables, analyzes question types, and separately prototypes model components. The notebooks are the executable product; the README documents their intended order and outputs, but no source excerpts were available to verify notebook internals. The graph therefore shows only documented relationships and omits the unimplemented application, training, and evaluation pipeline.

Main entry points: [`notebooks/01_dataset_analysis.ipynb`](https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/01_dataset_analysis.ipynb) (Dataset inspection), [`notebooks/03_prepare_aqa_dataset.ipynb`](https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/03_prepare_aqa_dataset.ipynb) (Answer aggregation), [`notebooks/02_audio_exploration.ipynb`](https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/02_audio_exploration.ipynb) (Audio exploration), [`notebooks/04_question_type_analysis.ipynb`](https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/04_question_type_analysis.ipynb) (Question analysis)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_data["Dataset artifacts"]
  node_annotations[("AQA annotations")]
  node_audio[("Local audio")]
  node_grouped[("Grouped QA tables")]
end

subgraph group_research["Notebook research"]
  node_inspection["Dataset inspection"]
  node_audioexplore["Audio exploration"]
  node_preparation["Answer aggregation"]
  node_classification["Question analysis"]
end

subgraph group_model["Model prototyping"]
  node_modelsetup["Model setup"]
end

node_researcher(("Researcher"))

node_researcher -->|"runs"| node_inspection
node_researcher -->|"runs"| node_audioexplore
node_researcher -->|"runs"| node_preparation
node_researcher -->|"runs"| node_classification
node_researcher -->|"runs"| node_modelsetup
node_inspection -->|"inspects"| node_annotations
node_inspection -->|"checks files"| node_audio
node_audioexplore -->|"loads audio"| node_audio
node_preparation -->|"groups annotations"| node_annotations
node_preparation -->|"generates tables"| node_grouped
node_classification -->|"adds categories"| node_grouped

click node_annotations "https://github.com/adhi-1235/audio-context-layer/tree/master/data/raw"
click node_grouped "https://github.com/adhi-1235/audio-context-layer/tree/master/data/processed"
click node_inspection "https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/01_dataset_analysis.ipynb"
click node_audioexplore "https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/02_audio_exploration.ipynb"
click node_preparation "https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/03_prepare_aqa_dataset.ipynb"
click node_classification "https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/04_question_type_analysis.ipynb"
click node_modelsetup "https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/03_model_setup.ipynb"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_annotations,node_audio,node_grouped toneBlue
class node_inspection,node_audioexplore,node_preparation,node_classification toneAmber
class node_modelsetup toneMint
class node_researcher toneIndigo
```

## Components

### Dataset artifacts

- [AQA annotations](https://github.com/adhi-1235/audio-context-layer/tree/master/data/raw): `data/raw`
- **Local audio**
- [Grouped QA tables](https://github.com/adhi-1235/audio-context-layer/tree/master/data/processed): `data/processed`

### Notebook research

- [Dataset inspection](https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/01_dataset_analysis.ipynb): `notebooks/01_dataset_analysis.ipynb`
- [Audio exploration](https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/02_audio_exploration.ipynb): `notebooks/02_audio_exploration.ipynb`
- [Answer aggregation](https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/03_prepare_aqa_dataset.ipynb): `notebooks/03_prepare_aqa_dataset.ipynb`
- [Question analysis](https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/04_question_type_analysis.ipynb): `notebooks/04_question_type_analysis.ipynb`

### Model prototyping

- [Model setup](https://github.com/adhi-1235/audio-context-layer/blob/master/notebooks/03_model_setup.ipynb): `notebooks/03_model_setup.ipynb`

### Other components

- **Researcher**

## Connections

- Researcher → Dataset inspection: runs (evidence: `README.md`)
- Researcher → Audio exploration: runs (evidence: `README.md`)
- Researcher → Answer aggregation: runs (evidence: `README.md`)
- Researcher → Question analysis: runs (evidence: `README.md`)
- Researcher → Model setup: runs (evidence: `README.md`)
- Dataset inspection → AQA annotations: inspects (evidence: `README.md`)
- Dataset inspection → Local audio: checks files (evidence: `README.md`)
- Audio exploration → Local audio: loads audio (evidence: `README.md`)
- Answer aggregation → AQA annotations: groups annotations (evidence: `README.md`)
- Answer aggregation → Grouped QA tables: generates tables (evidence: `README.md`)
- Question analysis → Grouped QA tables: adds categories (evidence: `README.md`)
