# vraind/gigaamui architecture

> Architecture diagram of the GitHub repository vraind/gigaamui, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-05). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/vraind/gigaamui
- Repository: https://github.com/vraind/gigaamui

## Overview

GigaAM UI Fixed is a Flask-based audio transcription app. Users upload audio through the web UI; the app queues a background job, normalizes audio, loads GigaAM on demand, transcribes short audio directly or long audio in chunks, saves TXT/JSON/SRT outputs, and exposes job status and downloads. Batch processing reuses the transcription workflow, while info and cleanup endpoints provide supporting controls. Evidence is limited to the README and sampled app.py excerpts; template behavior and unsampled route details are not inferred.

Main entry points: [`app.py`](https://github.com/vraind/gigaamui/blob/main/app.py) (Flask routes), [`templates/index.html`](https://github.com/vraind/gigaamui/blob/main/templates/index.html) (Web UI)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_web["Web interface"]
  node_ui["Web UI<br/>[index.html]"]
  node_flask["Flask routes<br/>[app.py]"]
end

subgraph group_jobs["Transcription jobs"]
  node_job_state[("Job state<br/>[app.py]")]
  node_batch["Batch processing<br/>[app.py]"]
  node_audio_prep["Audio preparation<br/>[app.py]"]
  node_model["GigaAM model<br/>[app.py]"]
  node_chunking["Long-audio chunking<br/>[app.py]"]
end

subgraph group_outputs["Results and files"]
  node_results["Output generation<br/>[app.py]"]
  node_audio_files[("Input audio files<br/>[app.py]")]
  node_saved_files[("TXT JSON SRT files<br/>[app.py]")]
end

node_user(("User"))
node_gigaam["GigaAM package"]
node_ffmpeg["FFmpeg"]

node_user -->|"uses"| node_ui
node_ui -.->|"requests"| node_flask
node_flask -->|"saves uploads"| node_audio_files
node_flask -->|"creates jobs"| node_job_state
node_flask -->|"starts batch"| node_batch
node_batch -->|"reads files"| node_audio_files
node_batch -->|"updates progress"| node_job_state
node_batch -->|"calls"| node_audio_prep
node_flask -->|"starts worker"| node_audio_prep
node_audio_prep -->|"normalizes via"| node_ffmpeg
node_audio_prep -->|"loads model"| node_model
node_model -->|"loads and transcribes"| node_gigaam
node_audio_prep -->|"chunks long audio"| node_chunking
node_chunking -->|"transcribes chunks"| node_model
node_audio_prep -->|"saves transcript"| node_results
node_results -->|"writes formats"| node_saved_files
node_flask -->|"serves status"| node_job_state
node_flask -->|"serves downloads"| node_saved_files

click node_ui "https://github.com/vraind/gigaamui/blob/main/templates/index.html"
click node_flask "https://github.com/vraind/gigaamui/blob/main/app.py"
click node_job_state "https://github.com/vraind/gigaamui/blob/main/app.py"
click node_batch "https://github.com/vraind/gigaamui/blob/main/app.py"
click node_audio_prep "https://github.com/vraind/gigaamui/blob/main/app.py"
click node_model "https://github.com/vraind/gigaamui/blob/main/app.py"
click node_chunking "https://github.com/vraind/gigaamui/blob/main/app.py"
click node_results "https://github.com/vraind/gigaamui/blob/main/app.py"
click node_audio_files "https://github.com/vraind/gigaamui/blob/main/app.py"
click node_saved_files "https://github.com/vraind/gigaamui/blob/main/app.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_ui,node_flask,node_user toneBlue
class node_job_state,node_batch,node_audio_prep,node_model,node_chunking toneAmber
class node_results,node_audio_files,node_saved_files toneMint
class node_gigaam,node_ffmpeg toneIndigo
```

## Components

### Web interface

- [Web UI](https://github.com/vraind/gigaamui/blob/main/templates/index.html): `templates/index.html`
- [Flask routes](https://github.com/vraind/gigaamui/blob/main/app.py): `app.py`

### Transcription jobs

- [Job state](https://github.com/vraind/gigaamui/blob/main/app.py): `app.py`
- [Batch processing](https://github.com/vraind/gigaamui/blob/main/app.py): `app.py`
- [Audio preparation](https://github.com/vraind/gigaamui/blob/main/app.py): `app.py`
- [GigaAM model](https://github.com/vraind/gigaamui/blob/main/app.py): `app.py`
- [Long-audio chunking](https://github.com/vraind/gigaamui/blob/main/app.py): `app.py`

### Results and files

- [Output generation](https://github.com/vraind/gigaamui/blob/main/app.py): `app.py`
- [Input audio files](https://github.com/vraind/gigaamui/blob/main/app.py): `app.py`
- [TXT JSON SRT files](https://github.com/vraind/gigaamui/blob/main/app.py): `app.py`

### Other components

- **User**
- **GigaAM package**
- **FFmpeg**

## Connections

- User → Web UI: uses (evidence: `README.md`)
- Web UI → Flask routes: requests
- Flask routes → Input audio files: saves uploads (evidence: `app.py`)
- Flask routes → Job state: creates jobs (evidence: `app.py`)
- Flask routes → Batch processing: starts batch (evidence: `app.py`)
- Batch processing → Input audio files: reads files (evidence: `app.py`)
- Batch processing → Job state: updates progress (evidence: `app.py`)
- Batch processing → Audio preparation: calls (evidence: `app.py`)
- Flask routes → Audio preparation: starts worker (evidence: `app.py`)
- Audio preparation → FFmpeg: normalizes via (evidence: `app.py`)
- Audio preparation → GigaAM model: loads model (evidence: `app.py`)
- GigaAM model → GigaAM package: loads and transcribes (evidence: `app.py`)
- Audio preparation → Long-audio chunking: chunks long audio (evidence: `app.py`)
- Long-audio chunking → GigaAM model: transcribes chunks (evidence: `app.py`)
- Audio preparation → Output generation: saves transcript (evidence: `app.py`)
- Output generation → TXT JSON SRT files: writes formats (evidence: `app.py`)
- Flask routes → Job state: serves status (evidence: `app.py`)
- Flask routes → TXT JSON SRT files: serves downloads (evidence: `app.py`)
