Speaker 1: Hello, in this video I want to share how I've supercharged my use of Obsidian, the knowledge management platform, with the help of OpenAI's Whisper platform and a community plugin. It's amazing to be able to speak to my mobile phone and get accurately transcribed voice notes into text, and that really helps me keep an active journal as well as take notes on the fly. So let's take a look at how that works. I'll open up Obsidian, which I'm running by the way in BlueStacks, an Android emulator on my computer here, and let's go add the Whisper plugin. Go to settings and choose community plugins and look for Whisper. I'll pick this one by Nick Danilo and install it. Now note that as with all use of OpenAI's Whisper API, I do need to have an API key. So I need to register for this OpenAI developer platform and it does cost a tiny amount to run this. Something like less than a cent for a paragraph of text. Now that I've installed Whisper and enabled it, I'll go to options and first of all enter my OpenAI API key. I'll delete it after this video so nobody else can use it. There are a few other options that we could look at here. An endpoint that might change over time. A model that might change over time. You might have prompts. For example, adding correct spellings or maybe you'd ask the transcription to translate what you're saying. So you will speak one language and get the output in another language. You can specify the language of the input. So I'll put it as English, but you could put it as something else. And there are some handy functions where you can also save the recordings. So whenever you speak to Obsidian, do you want to save that audio file somewhere? I'll turn it off to save space. And do you want to save that transcription in a separate file? Well, I'll always be saving it in my daily note or my impromptu note. So I'll put these off. So all I needed to do was just have my OpenAI API key. And now I've got that enabled. So I can go to one of my test notes here. And then I can run my command palette. Or this here. And start recording this way. So hello, now I'm speaking to OpenAI's Whisper API. And I should get an accurate transcription right into Obsidian very easily. Woohoo. Then we'd toggle that off. Wonderful. The audio was parsed successfully. So this works now anywhere I want to use it in Obsidian. But you did see that this was quite clunky for me to go to the command palette on mobile. So how do I make this faster? I want to show you one more trick. I can go to the settings here again. And go to mobile. And configure my mobile quick action. So it's already at Whisper. But I can then look for my start and stop recording thing. What that means is when I'm in any note, I can just pull from the top. And start the recording instead. So I'm just pulling from the top of my app to trigger my quick action. Which I've configured to be this voice recording. So it's always there and always super handy to activate and finish. And once I'm done, I'll just pull it again. And there we go. So it's a simple thing. The right plugin. API key. And a little bit of a trick to make it more usable by adding it as a quick action. I'm sure you can figure that out by yourself. But I've just been having so much fun with it that I really wanted to share it with you. Alright. Well, thanks for watching. And don't forget to have an amazing day. Bye.