homelab-00/TranscriptionSuite

A fully local and private Speech-To-Text app, offering multiple model backends, diarization & calendar mode - Available for Windows, macOS & Linux

What it solves

TranscriptionSuite is a free, open-source application that allows users to convert speech to text locally on their own computer. It eliminates the need to send audio data to cloud services, ensuring privacy and security while providing accurate transcripts with speaker labels.

How it works

The project consists of two main components: a desktop Dashboard (built with Electron) and a transcription server (Python backend). The server can run natively on Apple Silicon via Metal/MLX or within Docker containers on Windows and Linux. It supports multiple speech-to-text backends including Whisper, NVIDIA NeMo, VibeVoice-ASR, SenseVoice, and whisper.cpp, with hardware acceleration for NVIDIA CUDA, Apple Metal, and AMD/Intel Vulkan.

Who it’s for

It is designed for anyone needing private, local transcription—such as students recording lectures, professionals dictating documents, or users who want to manage their audio notes in a searchable notebook with an integrated AI assistant.

Highlights

  • 100% Local & Private: Audio and transcripts never leave the machine.
  • Speaker Diarization: Automatic "who said what" labeling using PyAnnote, CAM++, Sortformer, or built-in model features.
  • Multi-Backend Support: Compatible with various models (Whisper, NeMo, SenseVoice, etc.) and hardware (NVIDIA, Apple Silicon, AMD/Intel).
  • Live Mode: Real-time, sentence-by-sentence transcription for continuous dictation.
  • Audio Notebook: A calendar-view storage system with full-text search and an AI assistant compatible with OpenAI-compatible providers.
  • Broad Compatibility: Supports 90+ languages via Whisper and bidirectional translation for European languages via Canary v2.

Related

  • Project
    kaixxx/noScribe

    A free, open-source application for local audio transcription and speaker identification, designed for researchers and journalists to ensure data privacy.

  • Project
    thewh1teagle/vibe

    A private, offline audio and video transcription tool that uses AI models like Whisper to convert speech to text locally on your device.

  • Project
    goodroot/hyprwhspr

    A native Linux speech-to-text tool providing fast, private, and system-wide dictation using local or cloud-based AI models.

  • Project
    OpenWhispr/openwhispr

    OpenWhispr is a cross‑platform desktop app for privacy‑first voice dictation, meeting transcription, and AI‑assistant interaction. It supports offline speech‑to‑text with Whisper‑based models or cloud services, offers speaker diarisation, note‑taking with semantic search, team collaboration, and a public API for developers.

  • Project
    watzon/pindrop

    A Mac-native AI dictation app that provides private, offline speech-to-text conversion using on-device transcription engines.