← Discover
Discover · GitHub · Transcribe.cpp↑ +395 today

Fast speech conversion on local systems

Transcribe.cpp is a speech-to-text inference library developed in C++ that supports more than 16 model families. Using the ggml infrastructure, this tool enables different audio processing models to run efficiently on local systems.

Updates

  • August 31, 2026: Stars 1,825 → 1,865, latest release v0.2.3 (August 30, 2026).
  • August 24, 2026: Stars 1,816 → 1,825, latest release v0.2.2 (August 24, 2026).
  • August 20, 2026: Stars 1,802 → 1,816, latest release v0.2.1 (August 20, 2026).
  • August 18, 2026: Stars 1,673 → 1,802, latest release v0.2.0 (August 17, 2026).

What you get

  • Support for 16 different model families
  • High performance on GPU and CPU
  • Efficient inference with GGUF format

Installation

Vulkan supported Linux installation
# Ubuntu/Debian
sudo apt install build-essential cmake libvulkan-dev glslc libopenblas-dev

cmake -B build -DTRANSCRIBE_VULKAN=ON
cmake --build build

If you don't write code

🤖 Paste this into your AI agent (Claude Code · Codex · Antigravity)

I want to convert a local audio file to text using the Transcribe.cpp tool. How can I process my 16 kHz mono WAV format audio file using the transcribe-cli tool compiled on my system and the GGUF format model file I downloaded? Please explain the command structure required for this process and the file paths I should pay attention to.

Related dictionary terms

Who it is forIt is for developers who want to run privacy-focused and fast speech recognition systems on their own hardware.
LicenseMIT

Links

TreScout did not build this tool · we found it in GitHub trends and wrote it up. This page describes the repository as of 2026-07-21: The star count and our text belong to that day, the repository may have changed since. Check the repository link for the current state. This page was machine-translated from the Turkish original · the Turkish version prevails.