GetTranscript

A local-first audio transcription toolkit. Download, transcribe, and polish — from YouTube video to publication-ready transcript in three commands. All speech recognition runs on your own GPU; audio never leaves your machine.
View source on GitHub →

Pipeline

🎬

YouTube / Audio

Any video or audio file

→

🎙

Whisper large-v3

GPU-accelerated STT

→

🤖

LLM Punctuation

LLM API formatting

→

📄

Polished Transcript

Clean, readable text

Features

Fully Local & Private

All speech recognition runs on your GPU. Only the text formatting step optionally calls an external API.

Batch Processing

Process entire folders of audio files with checkpoint support for pause and resume.

GPU Accelerated

Best with 8GB+ VRAM. Falls back to int8 for smaller GPUs, or CPU mode if no NVIDIA card.

Streaming Output

Watch tokens arrive in real time with streaming mode, or wait for the complete result.