Windows — Version 1.0

Download ParlaAI

Transcribe, translate, and analyze your files — privately, on your own machine.

ParlaAI-Installer.exe · ~77 MB · Windows 10 / 11

Download Trial

Getting Started

Three steps to get up and running

The installer handles everything — no technical knowledge required.

Download the Installer

Download the installer to your Windows computer

Run the installer

Double-click the installer and it will prompt you through the installation.

Launch from your Desktop

The installer creates a ParlaAI app on your desktop. Open ParlaAI.

System Requirements

What you need

The installer checks all of these and installs anything missing automatically.

🖥️

Operating System

Windows 10 or 11 (64-bit)

💾

RAM

8GB minimum, 16 GB+ recommended

💿

Disk Space

20 GB free for AI models

🌐

Internet

Required for setup only. Fully offline after installation.

⚡

GPU (optional)

NVIDIA GPU speeds up transcription. CPU-only mode also supported.

Your data never leaves your machine

ParlaAI runs entirely on your computer using locally stored AI models. No audio, documents, transcripts, or any other data are transmitted to any server — ever. After a one-time setup that downloads the AI models, ParlaAI works with no internet connection at all. It is purpose-built for environments where confidentiality is non-negotiable.

🔒

Optional Feature

Speaker Identification

Know who said what, automatically.

OPTIONAL

Identify speakers in transcripts

ParlaAI can label each speaker in a transcript (SPEAKER_01, SPEAKER_02, etc.), making it easy to follow multi-person recordings such as interviews or hearings. This feature uses the pyannote speaker diarization model, which requires a free HuggingFace account and acceptance of the model's terms of use.

The installer will prompt you to optionally enter a HuggingFace token during setup. You can skip it and enable speaker identification any time by re-running the installer.

Download Trial