# audio-editing-challenge/audio-editing-challenge-baseline architecture

> Architecture diagram of the GitHub repository audio-editing-challenge/audio-editing-challenge-baseline, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-06). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/audio-editing-challenge/audio-editing-challenge-baseline
- Repository: https://github.com/audio-editing-challenge/audio-editing-challenge-baseline

## Overview

This repository provides two MMAE audio-editing baselines: a single AuK model track and an LLM-routed agent track. Both take benchmark samples through a runner and produce edited audio and evaluator manifests. In the agent track, the router chooses and executes sequential DSP, separation, or generative-edit tools; the latter two use separate HTTP services. The graph includes documented capabilities whose implementation excerpts were not provided; those links are identified as documented rather than inferred from unsampled code.

Main entry points: [`src/agent/router.py`](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/router.py) (LLM router), [`run_agent_mmae.py`](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/run_agent_mmae.py) (Agent runner), [`run_single_mmae.py`](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/run_single_mmae.py) (Single runner), [`src/agent/tools/dsp.py`](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/tools/dsp.py) (DSP tools)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_tracks["Baseline tracks"]
  node_single_runner["Single runner<br/>[run_single_mmae.py]"]
  node_agent_runner["Agent runner<br/>[run_agent_mmae.py]"]
end

subgraph group_agent["Agent orchestration"]
  node_config["Agent config<br/>[config.py]"]
  node_router["LLM router<br/>[router.py]"]
end

subgraph group_editing["Audio editing"]
  node_dsp_tools["DSP tools<br/>[dsp.py]"]
  node_audio_utils["Audio utilities<br/>[audio_utils.py]"]
end

subgraph group_services["Model services"]
  node_sam_client["SAM-Audio client"]
  node_sam_server["SAM-Audio service"]
  node_auk_client["AuK client<br/>[auk_client.py]"]
  node_auk_server["AuK service<br/>[auk_server.py]"]
end

subgraph group_io["Benchmark I/O"]
  node_mmae_io["MMAE manifests<br/>[mmae_io.py]"]
  node_edited_audio["Edited audio"]
end

node_benchmark["MMAE inputs"]
node_auk_model{{"AuK base model"}}
node_llm{{"OpenAI-compatible LLM"}}
node_sam_model{{"SAM-Audio model"}}
node_auk_agent_model{{"AuK generative model"}}

node_benchmark -->|"provides samples"| node_single_runner
node_benchmark -->|"provides samples"| node_agent_runner
node_single_runner -.->|"edits audio"| node_auk_model
node_single_runner -->|"writes audio"| node_edited_audio
node_agent_runner -->|"builds config"| node_config
node_agent_runner -->|"routes samples"| node_router
node_agent_runner -->|"builds manifests"| node_mmae_io
node_single_runner -->|"builds manifests"| node_mmae_io
node_agent_runner -->|"stores output"| node_edited_audio
node_router -->|"reads settings"| node_config
node_router -->|"requests tool calls"| node_llm
node_router -->|"executes DSP"| node_dsp_tools
node_dsp_tools -->|"loads and saves"| node_audio_utils
node_router -->|"dispatches separation"| node_sam_client
node_sam_client -->|"posts audio"| node_sam_server
node_sam_server -.->|"runs separation"| node_sam_model
node_router -->|"dispatches edit"| node_auk_client
node_auk_client -->|"posts audio"| node_auk_server
node_auk_server -.->|"runs generation"| node_auk_agent_model

click node_single_runner "https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/run_single_mmae.py"
click node_agent_runner "https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/run_agent_mmae.py"
click node_config "https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/config.py"
click node_router "https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/router.py"
click node_dsp_tools "https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/tools/dsp.py"
click node_audio_utils "https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/audio_utils.py"
click node_sam_client "https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/clients/sam_audio_client.py"
click node_sam_server "https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/services/sam_audio_server.py"
click node_auk_client "https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/clients/auk_client.py"
click node_auk_server "https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/services/auk_server.py"
click node_mmae_io "https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/mmae_io.py"

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_single_runner,node_agent_runner toneBlue
class node_config,node_router toneAmber
class node_dsp_tools,node_audio_utils toneMint
class node_sam_client,node_sam_server,node_auk_client,node_auk_server toneRose
class node_mmae_io,node_edited_audio,node_benchmark,node_auk_model,node_llm,node_sam_model,node_auk_agent_model toneIndigo
```

## Components

### Baseline tracks

- [Single runner](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/run_single_mmae.py): `run_single_mmae.py`
- [Agent runner](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/run_agent_mmae.py): `run_agent_mmae.py`

### Agent orchestration

- [Agent config](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/config.py): `src/agent/config.py`
- [LLM router](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/router.py): `src/agent/router.py`

### Audio editing

- [DSP tools](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/tools/dsp.py): `src/agent/tools/dsp.py`
- [Audio utilities](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/audio_utils.py): `src/agent/audio_utils.py`

### Model services

- [SAM-Audio client](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/clients/sam_audio_client.py): `src/agent/clients/sam_audio_client.py`
- [SAM-Audio service](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/services/sam_audio_server.py): `src/services/sam_audio_server.py`
- [AuK client](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/agent/clients/auk_client.py): `src/agent/clients/auk_client.py`
- [AuK service](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/services/auk_server.py): `src/services/auk_server.py`

### Benchmark I/O

- [MMAE manifests](https://github.com/audio-editing-challenge/audio-editing-challenge-baseline/blob/main/src/mmae_io.py): `src/mmae_io.py`
- **Edited audio**

### Other components

- **MMAE inputs**
- **AuK base model**
- **OpenAI-compatible LLM**
- **SAM-Audio model**
- **AuK generative model**

## Connections

- MMAE inputs → Single runner: provides samples (evidence: `README.md`)
- MMAE inputs → Agent runner: provides samples (evidence: `README.md`)
- Single runner → AuK base model: edits audio (evidence: `README.md`)
- Single runner → Edited audio: writes audio (evidence: `README.md`)
- Agent runner → Agent config: builds config (evidence: `run_agent_mmae.py`)
- Agent runner → LLM router: routes samples (evidence: `run_agent_mmae.py`)
- Agent runner → MMAE manifests: builds manifests (evidence: `run_agent_mmae.py`)
- Single runner → MMAE manifests: builds manifests (evidence: `run_single_mmae.py`)
- Agent runner → Edited audio: stores output (evidence: `run_agent_mmae.py`)
- LLM router → Agent config: reads settings (evidence: `src/agent/router.py`)
- LLM router → OpenAI-compatible LLM: requests tool calls (evidence: `src/agent/router.py`)
- LLM router → DSP tools: executes DSP (evidence: `src/agent/router.py`)
- DSP tools → Audio utilities: loads and saves (evidence: `src/agent/tools/dsp.py`)
- LLM router → SAM-Audio client: dispatches separation (evidence: `src/agent/router.py`)
- SAM-Audio client → SAM-Audio service: posts audio (evidence: `src/agent/clients/sam_audio_client.py`)
- SAM-Audio service → SAM-Audio model: runs separation (evidence: `README.md`)
- LLM router → AuK client: dispatches edit (evidence: `src/agent/router.py`)
- AuK client → AuK service: posts audio (evidence: `src/agent/clients/auk_client.py`)
- AuK service → AuK generative model: runs generation (evidence: `README.md`)
