Intervu Releases

Binary releases for Intervu — Real-time interview assistance powered by AI, running on your own infrastructure.

Download

Get the latest version from Releases.

Windows

• Intervu-Setup-x.x.x.exe — Installer
• Intervu-x.x.x-portable.exe — Portable (no installation required)

macOS

• Intervu-x.x.x-arm64.dmg — Apple Silicon (M1/M2/M3/M4)
• Intervu-x.x.x.dmg — Intel Macs
• Intervu-x.x.x-arm64-mac.zip — Apple Silicon (portable)
• Intervu-x.x.x-mac.zip — Intel Macs (portable)

:::warning macOS Users The macOS app is not code-signed. On first launch, remove the quarantine attribute:

xattr -d com.apple.quarantine /Applications/Intervu.app

:::

System Requirements

Windows

• OS: Windows 10 or later (64-bit)
• RAM: 4GB minimum, 8GB recommended
• Storage: ~500MB

macOS

• OS: macOS 12 (Monterey) or later
• RAM: 4GB minimum, 8GB recommended
• Storage: ~500MB
• FFmpeg: Must be installed via Homebrew (brew install ffmpeg)

Prerequisites

Before using Intervu, you need:

1. Virtual Audio Device — For capturing system audio

   • Windows: VB-Cable — Install as administrator and reboot
   • macOS: BlackHole — Configure multi-output device in Audio MIDI Setup

2. STT Endpoint — Choose one:

   • Speaches (recommended for standard transcription) — OpenAI-compatible, Installation Guide
     • Model: Systran/faster-distil-whisper-small.en
   • WhisperX ASR Service (required for speaker diarization) — Returns speaker embeddings for multi-speaker tracking
     • See Speaker Diarization docs for setup

3. LLM Endpoint — Choose one:

   • Ollama (recommended): ollama pull llama3.2:3b && ollama serve
   • OpenWebUI: Your private instance
   • OpenAI API: Requires API key

Quick Start

1. Install virtual audio device (VB-Cable on Windows, BlackHole on macOS) and reboot
2. Start Speaches for STT:

   docker run -d --gpus all -p 8000:8000 -v hf-hub-cache:/home/ubuntu/.cache/huggingface/hub ghcr.io/speaches-ai/speaches:latest-cuda

3. Start Ollama for LLM:

   ollama pull llama3.2:3b
   ollama serve

4. Configure Intervu Settings:
   • System Audio: CABLE Output (Windows) or BlackHole multi-output (macOS)
   • Microphone: Your microphone device
   • STT Endpoint: http://localhost:8000/v1/audio/transcriptions
   • STT Model: Systran/faster-distil-whisper-small.en
   • LLM Endpoint: http://localhost:11434/v1/chat/completions
   • LLM Model: llama3.2:3b

For Panel Interviews (Speaker Diarization)

If you need multi-speaker detection, use WhisperX ASR Service instead of Speaches:

1. Create a .env file with your HuggingFace token (requires access to pyannote/speaker-diarization-3.1, pyannote/segmentation-3.0, and pyannote/speaker-diarization-community-1):

   HF_TOKEN=your_huggingface_token_here
   ASR_MODEL=large-v3
   ASR_ENGINE=faster_whisper

2. Deploy the container:

   docker compose up -d

3. In Intervu Settings, set STT Endpoint to http://localhost:9000/asr and enable Speaker Diarization

See the Speaker Diarization guide for full setup instructions.

Documentation

Full documentation available at intervu-docs.aldrickb.app:

• Installation Guide
• macOS Setup
• Audio Configuration
• Speaker Diarization
• Interview Timer
• Troubleshooting
• Changelog
• Roadmap

Features

• Cross-platform — Windows and macOS support
• Real-time audio capture from system audio and microphone with speaker attribution
• Speech-to-text transcription using Whisper via Speaches or WhisperX ASR Service
• AI-powered answers using your own LLM (Ollama, OpenWebUI, OpenAI, or any OpenAI-compatible endpoint)
• Speaker Diarization — Multi-speaker identification for panel interviews with AI-assisted naming
• Interview Timer — Track interview duration and speaking ratios with time warnings
• Advanced Mode — Dual-LLM question extraction and intelligent answer queue
• Answer quality feedback — Rate answers with thumbs up/down for in-context learning
• Contextual answer tips — Key points, confidence scores, and structured format
• Q&A Bank & Context — Pre-prepared answers, company/role context for tailored responses
• Floating overlay — Always-on-top compact dark theme
• Privacy-first — All processing runs on your infrastructure, no data leaves your machine

Support

• Issues: GitHub Issues
• Docs: intervu-docs.aldrickb.app
• Website: aldrickb.com

Credits

Developer

• Aldrick Bonaobra — aldrickb.com

Technologies

• Electron — Cross-platform desktop framework
• React — UI framework
• Vite — Build tool
• Zustand — State management
• shadcn/ui — UI components
• Tailwind CSS — Styling

Dependencies

• FFmpeg — Audio capture and processing
• Speaches — Speech-to-text (OpenAI-compatible)
• WhisperX ASR Service — Speech-to-text with diarization (optional)
• Ollama — Local LLM inference (optional)

Special Thanks

• VB-Audio — VB-Cable virtual audio device (Windows)
• BlackHole — Virtual audio driver (macOS)

License

Copyright © 2026 Aldrick Bonaobra. All rights reserved.

This software is provided free of charge for personal and commercial use. The source code is not publicly available. Redistribution, modification, or reverse engineering is not permitted without written permission.