OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Audio & Voice AI

Aria

Aria is a Windows desktop application that captures system or microphone audio, displays recognized speech in a movable subtitle overlay, and can show translated text in a separate overlay.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Aria — OpenViST project profile in Audio & Voice AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Aria when you want a movable text overlay for Windows audio, or a separate translated overlay. Choose Lite for the Windows Live Captions workflow; choose Full for additional recognition modes, microphone capture, or bundled offline models.

Who it is for

  • Windows users who want captions for audio from videos, browsers, games or calls.
  • People who want a separate translated-caption overlay while listening.
  • Users who need a choice between Windows Live Captions and additional speech-recognition modes, including local models.

Documented uses

  • Turn computer playback or microphone audio into live subtitles.
  • Display a translation separately from the original captions.
  • Choose recognition suited to recorded material, fast-changing conversations, or Windows Live Captions.
Before you choose

Requirements & limits

The project is published under GPL-3.0. Lite requires Windows 11 version 22H2 or later; Full requires Windows 10 or 11. The packages include a Python runtime.

  • Lite is approximately 600 MB and offers Windows Live Captions mode; Full is approximately 7.6 GB and adds Precise and Realtime modes plus offline models.
  • Full supports system-audio or microphone capture; Lite is limited to Live Captions.
  • Precise uses Faster-Whisper and works best with an NVIDIA GPU; Live Captions requires no NVIDIA GPU or CUDA setup.
  • Realtime uses Sherpa-ONNX for Chinese and English or Vosk for Japanese.
  • Translation can use online services or the Full package’s local NLLB model; online services may be rate-limited or change.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install and start ARIA Full for live subtitles

Extract the Full package and start ARIA with a selected speech-recognition mode.

Before you start

  • Windows 10 or Windows 11
  • ARIA Full package; an NVIDIA GPU is recommended for Precise mode

Steps

  1. Download and extract the Full package.

  2. Start ARIA.bat from the extracted package.

  3. Select a recognition mode, configure translation if needed, and select Start.

Expected outcome

ARIA starts with the selected recognition mode and can display recognized speech in a movable subtitle overlay.

Guide limitations

  • The Full package is approximately 7.6 GB.
  • Precise mode works best with an NVIDIA GPU.
  • Online translation services may be rate-limited or change without notice.
Documentation 1 ↗
Source-based practical guide

Install and start ARIA Lite

Extract and start the Lite build for Windows Live Captions subtitles.

Before you start

  • Windows 11, version 22H2 or later
  • ARIA Lite package

Steps

  1. Download and extract the Lite package.

  2. Start run_lite.bat from the extracted package.

Expected outcome

ARIA Lite starts using Windows Live Captions and includes a translation overlay.

Guide limitations

  • The Lite build is limited to Windows Live Captions mode.
  • The Lite package is approximately 600 MB.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
sayksii/Aria · Official homepage title · retained in its source language
How we review sourcesSuggest a correction