How to Transcribe YouTube Videos on Mac
Whether you are a content creator repurposing video into blog posts, a student taking notes from lectures, or a researcher analyzing interviews, having a text version of YouTube videos is incredibly useful. YouTube auto-captions exist but are often inaccurate and hard to export. Here is how to get a clean, accurate transcript from any YouTube video playing on your Mac.
Step-by-Step Guide
Open the YouTube video in your browser
Navigate to the YouTube video you want to transcribe. You can use any browser — Safari, Chrome, Firefox, or Arc. Make sure your Mac's audio output is working and the video volume is at a reasonable level (not muted).
Set up system audio capture
Open Glasscribe from the menu bar and select "System Audio" as the input source. This captures the audio output from your Mac, so it will pick up whatever the YouTube video is playing. No browser extensions or virtual audio drivers are required.
Select the correct language
Set the transcription language to match the language spoken in the video. If you are watching a foreign-language video, you can enable live translation to get a real-time translation in your preferred language alongside the original transcription.
Play the video and start transcribing
Start the transcription, then play the YouTube video. Words will appear in real time as the video plays. You can use the floating overlay to watch the transcript build while the video plays. For long videos, you can pause and resume both the video and transcription as needed.
Export your transcript
When the video ends, stop the transcription. Export as .txt for a clean text transcript, or as .srt if you want timestamped subtitles. The .srt format is especially useful if you plan to add accurate captions to your own videos or create subtitle files.
Pro Tips
Frequently Asked Questions
Can I transcribe YouTube videos that do not have captions?
Yes. Unlike YouTube's built-in caption feature, system audio transcription works with any video regardless of whether the uploader has enabled captions. As long as there is audible speech, you can generate a transcript.
Is it faster to use YouTube's built-in transcript?
YouTube's auto-generated transcript (available under the "..." menu below a video) is instant but often contains errors and lacks proper punctuation. Transcribing the audio yourself produces a more accurate result, and you get a properly formatted export file.
Can I transcribe a YouTube video in one language and get the text in another?
Yes. With live translation enabled, you can transcribe a video in its original language and see a real-time translation in your preferred language. This works entirely on-device and supports 22+ language pairs.