Quick Guide: Transcribing Audio with WhisperAI (Full Transcript)

Speaker 1: Hey guys, so in this video, I am gonna show you how you can use OpenAI new AI model called WhisperAI in 60 seconds. It is used to transcribe your audio file. And it is super duper easy. So first of all let's start with installing OpenAI package using pip install OpenAI or if you already have it installed, upgrade it. As you can see it is installed now. Create a file called main.py in your favorite code editor. Now let's start with importing OpenAI package. Just type OpenAI and then we need an API key for this. So for that, just go to the website called platform.openai.com then click on your profile pic, then view API keys. And then just generate the new secret key and just copy it. Now go to your code editor and do what I do. Here we are simply using .api key attribute from OpenAI to set up API key. And now here we are just loading the music. The open function is commonly used in Python to read files. And here rb stands for read binary which simply specifies that the file should be opened in binary mode. And I had this audio file so I just loaded that. Now let's transcribe this audio. This line uses the transcribe method of the OpenAI audio class to convert the audio into text. We are using the whisper model to transcribe file and simply passing object pointing to the audio file that is audio underscore file. And I am simply printing it and now let's see what we get. See I am getting the transcript of the audio file. The audio was just 20 seconds long. But you can transcript the large files too and if your file is large that 25mb then you can spilt it using pydub and then load it. So thank you for now and I will see you on the next one.