# sushii-12/voxfusion architecture

> Architecture diagram of the GitHub repository sushii-12/voxfusion, 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/sushii-12/voxfusion
- Repository: https://github.com/sushii-12/voxfusion

## Overview

VoxFusion analyzes uploaded or recorded audio for family-emergency voice-cloning scam signals. The web workflow manages family members and reference/sample audio, then runs either an AASIST-only baseline or a combined analysis that gathers speaker, deepfake, transcription, and intent evidence before fusion returns risk and reasons. The README also describes GSM/telephone-codec evaluation. Source excerpts establish the main orchestration and adapter links; some endpoint and storage details are documented but not fully sampled.

Main entry points: [`frontend/backend/app/services`](https://github.com/sushii-12/voxfusion/tree/main/frontend/backend/app/services) (Analysis adapters), [`frontend/backend/app/main.py`](https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/main.py) (FastAPI API), [`frontend/backend/app/services/fusion.py`](https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/services/fusion.py) (Intent and fusion adapters), [`frontend/frontend/src/main.jsx`](https://github.com/sushii-12/voxfusion/blob/main/frontend/frontend/src/main.jsx) (React application)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_web["Web interface"]
  node_ui["React application<br/>[main.jsx]"]
end

subgraph group_backend["Application backend"]
  node_api["FastAPI API<br/>[main.py]"]
  node_models["Domain records<br/>[models.py]"]
  node_db[("SQLite and uploads<br/>[db.py]")]
  node_schemas["Result schemas<br/>[schemas.py]"]
  node_pipeline["Analysis adapters"]
end

subgraph group_signals["Detection signals"]
  node_aasist_adapter["AASIST adapter<br/>[aasist.py]"]
  node_ecapa_adapter["ECAPA adapter<br/>[ecapa.py]"]
  node_whisper_adapter["Whisper adapter<br/>[whisper.py]"]
  node_aasist["AASIST detection"]
  node_ecapa["Speaker verification"]
  node_transcription["Audio transcription<br/>[transcription.py]"]
  node_gsm["GSM codec evaluation"]
end

subgraph group_intent["Intent analysis"]
  node_intent_adapter["Intent and fusion adapters<br/>[fusion.py]"]
  node_intent_v1["Rule-based intent<br/>[scam_intent.py]"]
  node_intent_v2["ML intent classifier<br/>[scam_intent_v2.py]"]
end

subgraph group_fusion["Risk fusion"]
  node_fusion_engine["Interpretable fusion<br/>[fusion.py]"]
end

node_user(("Analyst / user"))

node_user -->|"uses"| node_ui
node_ui -->|"calls API"| node_api
node_api -->|"queries records"| node_models
node_api -->|"uses sessions"| node_db
node_api -->|"runs analysis"| node_pipeline
node_pipeline -->|"invokes"| node_aasist_adapter
node_pipeline -->|"invokes"| node_ecapa_adapter
node_pipeline -->|"invokes"| node_whisper_adapter
node_aasist_adapter -->|"delegates to"| node_aasist
node_ecapa_adapter -->|"delegates to"| node_ecapa
node_whisper_adapter -->|"delegates to"| node_transcription
node_pipeline -->|"combines signals"| node_intent_adapter
node_intent_adapter -->|"analyzes transcript"| node_intent_v1
node_intent_adapter -->|"analyzes transcript"| node_intent_v2
node_intent_adapter -->|"delegates fusion"| node_fusion_engine
node_pipeline -->|"returns typed results"| node_schemas
node_gsm -.->|"evaluates codec audio"| node_aasist

click node_ui "https://github.com/sushii-12/voxfusion/blob/main/frontend/frontend/src/main.jsx"
click node_api "https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/main.py"
click node_models "https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/models.py"
click node_db "https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/db.py"
click node_schemas "https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/services/schemas.py"
click node_pipeline "https://github.com/sushii-12/voxfusion/tree/main/frontend/backend/app/services"
click node_aasist_adapter "https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/services/aasist.py"
click node_ecapa_adapter "https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/services/ecapa.py"
click node_whisper_adapter "https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/services/whisper.py"
click node_aasist "https://github.com/sushii-12/voxfusion/blob/main/src/deepfake_detection.py"
click node_ecapa "https://github.com/sushii-12/voxfusion/blob/main/src/speaker_verification.py"
click node_transcription "https://github.com/sushii-12/voxfusion/blob/main/src/transcription.py"
click node_intent_adapter "https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/services/fusion.py"
click node_intent_v1 "https://github.com/sushii-12/voxfusion/blob/main/src/scam_intent.py"
click node_intent_v2 "https://github.com/sushii-12/voxfusion/blob/main/src/scam_intent_v2.py"
click node_fusion_engine "https://github.com/sushii-12/voxfusion/blob/main/src/fusion.py"
click node_gsm "https://github.com/sushii-12/voxfusion/tree/main/scripts"

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_ui,node_user toneBlue
class node_api,node_models,node_db,node_schemas,node_pipeline toneAmber
class node_aasist_adapter,node_ecapa_adapter,node_whisper_adapter,node_aasist,node_ecapa,node_transcription,node_gsm toneMint
class node_intent_adapter,node_intent_v1,node_intent_v2 toneRose
class node_fusion_engine toneIndigo
```

## Components

### Web interface

- [React application](https://github.com/sushii-12/voxfusion/blob/main/frontend/frontend/src/main.jsx): `frontend/frontend/src/main.jsx`

### Application backend

- [FastAPI API](https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/main.py): `frontend/backend/app/main.py`
- [Domain records](https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/models.py): `frontend/backend/app/models.py`
- [SQLite and uploads](https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/db.py): `frontend/backend/app/db.py`
- [Result schemas](https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/services/schemas.py): `frontend/backend/app/services/schemas.py`
- [Analysis adapters](https://github.com/sushii-12/voxfusion/tree/main/frontend/backend/app/services): `frontend/backend/app/services`

### Detection signals

- [AASIST adapter](https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/services/aasist.py): `frontend/backend/app/services/aasist.py`
- [ECAPA adapter](https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/services/ecapa.py): `frontend/backend/app/services/ecapa.py`
- [Whisper adapter](https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/services/whisper.py): `frontend/backend/app/services/whisper.py`
- [AASIST detection](https://github.com/sushii-12/voxfusion/blob/main/src/deepfake_detection.py): `src/deepfake_detection.py`
- [Speaker verification](https://github.com/sushii-12/voxfusion/blob/main/src/speaker_verification.py): `src/speaker_verification.py`
- [Audio transcription](https://github.com/sushii-12/voxfusion/blob/main/src/transcription.py): `src/transcription.py`
- [GSM codec evaluation](https://github.com/sushii-12/voxfusion/tree/main/scripts): `scripts`

### Intent analysis

- [Intent and fusion adapters](https://github.com/sushii-12/voxfusion/blob/main/frontend/backend/app/services/fusion.py): `frontend/backend/app/services/fusion.py`
- [Rule-based intent](https://github.com/sushii-12/voxfusion/blob/main/src/scam_intent.py): `src/scam_intent.py`
- [ML intent classifier](https://github.com/sushii-12/voxfusion/blob/main/src/scam_intent_v2.py): `src/scam_intent_v2.py`

### Risk fusion

- [Interpretable fusion](https://github.com/sushii-12/voxfusion/blob/main/src/fusion.py): `src/fusion.py`

### Other components

- **Analyst / user**

## Connections

- Analyst / user → React application: uses (evidence: `README.md`)
- React application → FastAPI API: calls API (evidence: `frontend/frontend/src/main.jsx`)
- FastAPI API → Domain records: queries records (evidence: `frontend/backend/app/main.py`)
- FastAPI API → SQLite and uploads: uses sessions (evidence: `frontend/backend/app/main.py`)
- FastAPI API → Analysis adapters: runs analysis (evidence: `frontend/backend/app/main.py`)
- Analysis adapters → AASIST adapter: invokes (evidence: `frontend/backend/app/main.py`)
- Analysis adapters → ECAPA adapter: invokes (evidence: `frontend/backend/app/main.py`)
- Analysis adapters → Whisper adapter: invokes (evidence: `frontend/backend/app/main.py`)
- AASIST adapter → AASIST detection: delegates to (evidence: `frontend/backend/app/services/aasist.py`)
- ECAPA adapter → Speaker verification: delegates to (evidence: `frontend/backend/app/services/ecapa.py`)
- Whisper adapter → Audio transcription: delegates to (evidence: `frontend/backend/app/services/whisper.py`)
- Analysis adapters → Intent and fusion adapters: combines signals (evidence: `frontend/backend/app/main.py`)
- Intent and fusion adapters → Rule-based intent: analyzes transcript (evidence: `frontend/backend/app/services/fusion.py`)
- Intent and fusion adapters → ML intent classifier: analyzes transcript (evidence: `frontend/backend/app/services/fusion.py`)
- Intent and fusion adapters → Interpretable fusion: delegates fusion (evidence: `frontend/backend/app/services/fusion.py`)
- Analysis adapters → Result schemas: returns typed results (evidence: `frontend/backend/app/services/aasist.py`)
- GSM codec evaluation → AASIST detection: evaluates codec audio (evidence: `README.md`)
