Guide to Transcribing Audio Using Whisper AI
Learn how to convert audio to text with Whisper AI via Google Collab setup. Perfect for content creators seeking quick and accurate transcriptions.
File
WHISPER AI
Added on 01/29/2025
Speakers
add Add new speaker

Speaker 1: Hello people, greetings from wherever you're watching right now. In this video, I'll show you how to transcribe audio to text using Whisper AI, the cutting edge technology by OpenAI. I'll show you step by step how to harness the power of Whisper AI to effortlessly convert spoken words into written text. Whether you're a content creator, journalist, or just looking to save time, this is a game changer. I am Joe, your tech handyman. Let's dive in.

Speaker 2: Okay guys, head over to Google Drive, this is the URL, drive.google.com when you're signed in. So here on the plus button that has new, click on it. From the drop down, go all the way to more, and then go down to connect more apps, and then this particular pop-up will come up this window. So on the search bar, just type in Google Collaboratory, like so, and click enter. Then these apps will come up. This is the particular app we want to install. Some people call it Google Collab. So just click on it and click on install. Click on continue, like that. Now it will ask you to choose the particular account that you want to install it on. You select that and click okay, okay, and here it will tell you that it's going to be connected to your Google Drive. So you click okay, click done, and then that's for that. So close this particular window. Now if you go back to that new button and click on it from the drop down, go down to more. From the next drop down, you can see Google Collaboratory. So you click on it, it takes you right here. This is where we are going to do whatever we want to do. So here, first thing, you want to change the filename. So once you change the filename, in this case it's audio to text, leave the extension IPNY. So here, click on the runtime, scroll all the way down to change runtime type. So in my own case, the default is Python 3 because this program is actually written in Python. So ensure the CPU is selected. So click okay. So in this box, you are going to paste in these two lines of code. The first one is to install Whisper through JIT and the second one is to install FFmpeg. So once done, just click on the play button and it installs. Let me just open this section so we can see what's up on there. So here, you click on this folder icon and then you want to bring in the folder. So you click on here, go to your system, select this folder that has the audio that you want to transcribe to text and then it will show you that the files are saved somewhere else because after the runtime, it will be removed. So you click okay and then this is a particular file that I'm going to do. So click on the code right here and then in this space, you are going to paste this one line of code again. This has to do with the Whisper prompt. So here, you are going to enter the file name. The file name has to be the exact same file name that you brought in the audio file that you want to transcribe to text. So I'm going to type joe.mp3 which is the exact same name here. So at the end, there are various models. We are not going to go into that. I'll just type in medium.n. So click on the play button and allow it to run. Once it gets to 100, you can see that there are other files right to the left. This one is mp3. This one is SRT. This one is TSV. The next one is TST and VTT. These are a bunch of files that contains different things. So if you open on the text one and then click on download, you download the text file and then open it like that. So it takes you to the file. You can see this is Joe. You take Handerman. Remember to subscribe to my channel. This is the exact same thing that is in the audio file. Now, if you open the SRT file and download it, and then again, you click open to open to see what's inside of it. You see that it's the same thing, but here you have the timestamp on it. This translation is very accurate. This transcription, should I say, is very accurate. So that is why it is proper, if you can, to use this for your subtitles. So as it stands, guys, now if you look at the entire file, the left side of things right here, once done, if you don't see this file, just click on this refresh button and all the files will come on board. So guys, thanks for watching.

ai AI Insights
Summary

Generate a brief summary highlighting the main points of the transcript.

Generate
Title

Generate a concise and relevant title for the transcript based on the main themes and content discussed.

Generate
Keywords

Identify and highlight the key words or phrases most relevant to the content of the transcript.

Generate
Enter your query
Sentiments

Analyze the emotional tone of the transcript to determine whether the sentiment is positive, negative, or neutral.

Generate
Quizzes

Create interactive quizzes based on the content of the transcript to test comprehension or engage users.

Generate
{{ secondsToHumanTime(time) }}
Back
Forward
{{ Math.round(speed * 100) / 100 }}x
{{ secondsToHumanTime(duration) }}
close
New speaker
Add speaker
close
Edit speaker
Save changes
close
Share Transcript