# dasilva333/airi-audio-server architecture

> Architecture diagram of the GitHub repository dasilva333/airi-audio-server, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-03). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/dasilva333/airi-audio-server
- Repository: https://github.com/dasilva333/airi-audio-server

## Overview

AIRI Audio Server is a Node.js API for local C++ audio inference. Clients send speech, voice, music, or sound-effect requests to the HTTP server; routes dispatch work to the relevant engines and return audio, metadata, or streaming responses. Voice registration adds normalization and optional transcription, while a shared GPU queue coordinates GPU tasks. Model setup/download is a separate provisioning capability. Relationships for sampled code are cited; broader capabilities and queue behavior are included where documented in the README.

Main entry points: [`src/server.js`](https://github.com/dasilva333/airi-audio-server/blob/master/src/server.js) (HTTP Server), [`src/routes.js`](https://github.com/dasilva333/airi-audio-server/blob/master/src/routes.js) (API Routes), [`src/voices.js`](https://github.com/dasilva333/airi-audio-server/blob/master/src/voices.js) (Voice Catalog), [`src/gpu.js`](https://github.com/dasilva333/airi-audio-server/blob/master/src/gpu.js) (GPU Binary Resolver)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_api["API and Coordination"]
  node_server["HTTP Server<br/>[server.js]"]
  node_routes["API Routes<br/>[routes.js]"]
  node_queue["GPU Queue<br/>[queue.js]"]
  node_text["Text Processing<br/>[text.js]"]
end

subgraph group_speech["Speech and Voices"]
  node_engine["Speech Engine<br/>[engine.js]"]
  node_voices[("Voice Catalog<br/>[voices.js]")]
  node_stt["Speech Recognition<br/>[stt.js]"]
  node_ffmpeg["Audio Normalization<br/>[ffmpeg.js]"]
end

subgraph group_creative["Audio Generation"]
  node_music["Music Engine<br/>[music.js]"]
  node_sfx["Sound Effects<br/>[sfx.js]"]
  node_voice_design["Voice Design<br/>[voice_design.js]"]
end

subgraph group_runtime["Inference Runtime"]
  node_gpu["GPU Binary Resolver<br/>[gpu.js]"]
  node_cpp["audio.cpp Inference"]
end

subgraph group_provision["Model Provisioning"]
  node_setup["Model Setup<br/>[setup.js]"]
end

node_client(("API Client"))
node_huggingface(("Hugging Face"))

node_client -->|"sends requests"| node_server
node_server -->|"mounts routes"| node_routes
node_server -->|"creates queue"| node_queue
node_server -->|"creates processor"| node_text
node_server -->|"creates engine"| node_engine
node_server -->|"creates manager"| node_voices
node_server -->|"creates engine"| node_music
node_server -->|"creates designer"| node_voice_design
node_server -->|"creates engine"| node_sfx
node_routes -->|"dispatches speech"| node_engine
node_routes -->|"registers voices"| node_voices
node_routes -->|"requests music"| node_music
node_routes -->|"converts uploads"| node_ffmpeg
node_routes -->|"transcribes audio"| node_stt
node_routes -->|"queues GPU tasks"| node_queue
node_engine -->|"resolves binary"| node_gpu
node_engine -->|"streams inference"| node_cpp
node_music -->|"resolves binary"| node_gpu
node_music -->|"runs inference"| node_cpp
node_sfx -->|"resolves binary"| node_gpu
node_voice_design -->|"resolves binary"| node_gpu
node_voice_design -->|"ingests generated voice"| node_voices
node_voices -->|"normalizes clips"| node_ffmpeg
node_voices -->|"transcribes references"| node_stt
node_stt -->|"normalizes input"| node_ffmpeg
node_stt -->|"resolves binary"| node_gpu
node_setup -->|"downloads weights"| node_huggingface

click node_server "https://github.com/dasilva333/airi-audio-server/blob/master/src/server.js"
click node_routes "https://github.com/dasilva333/airi-audio-server/blob/master/src/routes.js"
click node_queue "https://github.com/dasilva333/airi-audio-server/blob/master/src/queue.js"
click node_text "https://github.com/dasilva333/airi-audio-server/blob/master/src/text.js"
click node_engine "https://github.com/dasilva333/airi-audio-server/blob/master/src/engine.js"
click node_voices "https://github.com/dasilva333/airi-audio-server/blob/master/src/voices.js"
click node_stt "https://github.com/dasilva333/airi-audio-server/blob/master/src/stt.js"
click node_ffmpeg "https://github.com/dasilva333/airi-audio-server/blob/master/src/ffmpeg.js"
click node_music "https://github.com/dasilva333/airi-audio-server/blob/master/src/music.js"
click node_sfx "https://github.com/dasilva333/airi-audio-server/blob/master/src/sfx.js"
click node_voice_design "https://github.com/dasilva333/airi-audio-server/blob/master/src/voice_design.js"
click node_gpu "https://github.com/dasilva333/airi-audio-server/blob/master/src/gpu.js"
click node_setup "https://github.com/dasilva333/airi-audio-server/blob/master/setup.js"

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_server,node_routes,node_queue,node_text,node_client toneBlue
class node_engine,node_voices,node_stt,node_ffmpeg toneAmber
class node_music,node_sfx,node_voice_design toneMint
class node_gpu,node_cpp toneRose
class node_setup,node_huggingface toneIndigo
```

## Components

### API and Coordination

- [HTTP Server](https://github.com/dasilva333/airi-audio-server/blob/master/src/server.js): `src/server.js`
- [API Routes](https://github.com/dasilva333/airi-audio-server/blob/master/src/routes.js): `src/routes.js`
- [GPU Queue](https://github.com/dasilva333/airi-audio-server/blob/master/src/queue.js): `src/queue.js`
- [Text Processing](https://github.com/dasilva333/airi-audio-server/blob/master/src/text.js): `src/text.js`

### Speech and Voices

- [Speech Engine](https://github.com/dasilva333/airi-audio-server/blob/master/src/engine.js): `src/engine.js`
- [Voice Catalog](https://github.com/dasilva333/airi-audio-server/blob/master/src/voices.js): `src/voices.js`
- [Speech Recognition](https://github.com/dasilva333/airi-audio-server/blob/master/src/stt.js): `src/stt.js`
- [Audio Normalization](https://github.com/dasilva333/airi-audio-server/blob/master/src/ffmpeg.js): `src/ffmpeg.js`

### Audio Generation

- [Music Engine](https://github.com/dasilva333/airi-audio-server/blob/master/src/music.js): `src/music.js`
- [Sound Effects](https://github.com/dasilva333/airi-audio-server/blob/master/src/sfx.js): `src/sfx.js`
- [Voice Design](https://github.com/dasilva333/airi-audio-server/blob/master/src/voice_design.js): `src/voice_design.js`

### Inference Runtime

- [GPU Binary Resolver](https://github.com/dasilva333/airi-audio-server/blob/master/src/gpu.js): `src/gpu.js`
- **audio.cpp Inference**

### Model Provisioning

- [Model Setup](https://github.com/dasilva333/airi-audio-server/blob/master/setup.js): `setup.js`

### Other components

- **API Client**
- **Hugging Face**

## Connections

- API Client → HTTP Server: sends requests (evidence: `README.md`)
- HTTP Server → API Routes: mounts routes (evidence: `src/server.js`)
- HTTP Server → GPU Queue: creates queue (evidence: `src/server.js`)
- HTTP Server → Text Processing: creates processor (evidence: `src/server.js`)
- HTTP Server → Speech Engine: creates engine (evidence: `src/server.js`)
- HTTP Server → Voice Catalog: creates manager (evidence: `src/server.js`)
- HTTP Server → Music Engine: creates engine (evidence: `src/server.js`)
- HTTP Server → Voice Design: creates designer (evidence: `src/server.js`)
- HTTP Server → Sound Effects: creates engine (evidence: `src/server.js`)
- API Routes → Speech Engine: dispatches speech (evidence: `README.md`)
- API Routes → Voice Catalog: registers voices (evidence: `src/routes.js`)
- API Routes → Music Engine: requests music (evidence: `src/routes.js`)
- API Routes → Audio Normalization: converts uploads (evidence: `src/routes.js`)
- API Routes → Speech Recognition: transcribes audio (evidence: `src/routes.js`)
- API Routes → GPU Queue: queues GPU tasks (evidence: `README.md`)
- Speech Engine → GPU Binary Resolver: resolves binary (evidence: `src/engine.js`)
- Speech Engine → audio.cpp Inference: streams inference (evidence: `src/engine.js`)
- Music Engine → GPU Binary Resolver: resolves binary (evidence: `src/music.js`)
- Music Engine → audio.cpp Inference: runs inference (evidence: `src/music.js`)
- Sound Effects → GPU Binary Resolver: resolves binary (evidence: `src/sfx.js`)
- Voice Design → GPU Binary Resolver: resolves binary (evidence: `src/voice_design.js`)
- Voice Design → Voice Catalog: ingests generated voice (evidence: `src/voice_design.js`)
- Voice Catalog → Audio Normalization: normalizes clips (evidence: `src/voices.js`)
- Voice Catalog → Speech Recognition: transcribes references (evidence: `src/voices.js`)
- Speech Recognition → Audio Normalization: normalizes input (evidence: `src/stt.js`)
- Speech Recognition → GPU Binary Resolver: resolves binary (evidence: `src/stt.js`)
- Model Setup → Hugging Face: downloads weights (evidence: `README.md`)
