The Easy Guide to Using OpenAI’s Whisper Model to Transcribe Video and Audio

Published by

on

OpenAI recently released Whisper, an open source automatic speech recognition model that’s incredibly powerful. I’m biased (I’m the Science Communicator for OpenAI), but in my experience it’s better than any system or service I’ve ever used. Best of all, your can use it completely free, either by downloading it to your computer or by running it in the cloud on a free service like Google Colab.

A lot of writers, podcasters and YouTubers use transcription on a regular basis. The Whisper model can dramatically improve the quality of transcription. It’s punctuation and grammar are incredible, as well as its word recognition. Whisper can also work in many different languages.

If any of this is unfamiliar to you, don’t worry. I’ve made three videos that can help even a complete beginner new to coding, Jupyter notebooks and machine learning.

The key part of this is using a Python notebook (or called a Jupyter notebook), a special notebook that lets you run code inside blocks in a document that mixes code and text.

Google Colab is a platform that lets you run the code on a cloud server. Colab offers free GPUs on the free tier, which makes machine learning applications run much faster. All you need to get started with Colab is a Google account. You can learn more about how their notebooks work here.

To get started with OpenAI’s Whisper model you can watch any of the videos below where I take you through each step. You can also find them on my GitHub page.

📺 Whisper YouTube Transcription Python Notebook

This takes a YouTube video url and generates a high quality transcript you can upload to YouTube with autosync. 

Getting started video: https://youtu.be/kENRf82_RQs

Click here to go to the: Whisper YouTube notebook


🎙️ Whisper Google Drive Audio Transcription Python Notebook

This connects to your Google Drive and will batch process audio files uploaded to /WhisperAudio 

Getting started video: https://youtu.be/yVLhG4-7Sj4

Click here to go to the: Whisper Google Drive Audio notebook


📼 Whisper Google Drive Video Transcription Python Notebook

This connects to your Google Drive and will batch process video files uploaded to /WhisperVideo 

Getting started video: https://youtu.be/YGpYinji7II

Click here to go to the: Whisper Google Drive Video notebook