You can save a ton of time (and energy) by dictating instead of typing.

I use OpenAI’s Whisper API religiously now to dictate a lot of my essays. Given that many of my blog articles are just conversational in nature, it doesn’t require that much thinking versus just dictating and talking. I often use this instead as a way to drastically increase the WPM of my typing speed.

Whisper is state-of-the-art when it comes to its dictation. In fact, I would say it is 10 times better than Google or Apple’s dictation. It is so good that I often don’t even have to modify any of the punctuation, capitalization, or nouns. I don’t know what it’s souped up on, but it is incredibly, incredibly good. I would definitely highly recommend it.

How to Use it

As of 2023-03-17, I highly recommend using Whispering, my wrapper for the Whisper API. It is, in my opinion, much easier to use and abstracts much of the clicking that you will manually need to do in this workflow.

However, if you still prefer to try it out in the playground, navigate to beta.openai.com/playground

1. Click on the “Speech to text” Button in the Top Right

2. Click “Start recording”

3. Click “Stop recording” And you’re done! You will now see the text that was generated by Whisper and you can copy it into your clipboard or edit it appropriately.