Speaker 1: Whisper is an automatic speech recognition, like ASR, system. What sets it apart is that it's trained on a massive 680,000 hours of multilingual and multitask supervised data collected from the web. That's a lot of training, my friends. Because of this extensive training, Whisper is incredibly versatile. Here I am on my Google Colab. You can see right over here. The first thing you need to do is, let's just change the name. I can just call it as speech to text. So the first thing you need to do is you need to install the model. It's available here like https://github.com/.openai.whisper.git. So you need to pip install this model. Let me just zoom in a bit. Then you need to actually do like sudo apt update and you need to update and install like ffmpeg. And then you need to run this thing, okay? If you run this thing, it will download and install and the model will be available for you. Then you need to give a prompt like whisper. Then you need to provide a recording and you need to specify the model. The medium is actually the one that is most optimized and .en means that you are doing it in English. So that's pretty much it. So now what I can do is I can just move on to chat gpt and I can say provide me a one minute script for the topic benefits of generative AI, okay? Let's see what it will provide me with. Yeah, so that's the script you can see right over here. So now I will try to read the script, okay? I will record reading the script like in different styles. Mostly I will just move on with like the normal style. Then I will just move with like the fast one, then the slow one. Then I will just try to read it with an accent to see whether it will transcribe it efficiently or not. So let me open up my recorder. Yeah, this is my sound recorder. You can see right over here. Yeah, my recording is actually done. You see how I just changed different styles, okay? Let's save this thing and if I just move on to like show in folder. So here is my recording. So now I can just drag my recording and upload inside of my Google Colab, okay? I will just upload this thing in Google Colab. It will be done in a while. I can just delete these things. Yeah, my recording is uploaded. You can see right over here, okay? So let me just take the name of the file and I can change the file here, okay? Then I can just run this cell to see whether it's showing me the transcription or not. It will actually provide you with the timestamps as well. So you can like use the SRT file, whether you are like doing a video on YouTube or something like that. You can use it anywhere you want, okay? It will provide you with a basic text file as well. You can use that as well, okay? So now you see how amazing is this thing. So thank you so much for watching this video. That marks the end of this video and I hope you like this one and I shall catch you up in the next one. Till then, have a good day. Bye.