Each module in Machine Learning Crash Course begins with a short video lecture (generally 3–5 minutes) by a Google machine learning researcher. These videos introduce the machine learning concepts discussed and reinforced in the module’s documentation and exercises.

Videos are hosted on YouTube, where available. If you have problems running a course video, please submit a bug to us via the Send Feedback widget at the top of the course page.

Video Features

Navigating the video player

  • The narration of each video is synced to a series of companion presentation slides, with clickable links and text you can copy.
  • The Previous button  below the video navigates to the preceding slide and corresponding video.
  • The Next button  below the video navigates to the following slide and corresponding video.
  • The speed buttons below the video allows you to adjust playback speed (1x is the original recording speed; 1.5x is one-and-a-half-speed; 2x is double-speed).

Video transcript

  • The Transcript dropdown provides timestamped subtitles for the audio.
  • Click on any transcript line to navigate to that portion of the video.
  • Transcripts are searchable. Enter keywords or phrases in the Search box to locate them in the transcript.
  • Transcript subtitles are currently available in English, French, Korean, Spanish (Latin American), and Simplified Chinese. You can change the subtitle language via the language dropdown in the gray navigation bar.

Audio dubbing

  • Videos were originally recorded in English.
  • Audio dubbing of the videos is currently available in French, Korean, Spanish (Latin American), and Mandarin Chinese. To hear dubbing in another language, change the language setting of the course as described in Translations.

Note: Audio dubbing for video lectures was generated using machine learning techniques. Please help us refine our text-to-speech model; click SEND FEEDBACK at the top of a course page to submit bug reports and suggestions.

Video vs. Text

In the ML Concepts modules, both the videos and text documentation cover the same topics. Videos provide a high-level overview of key concepts, and the companion documentation provides a deeper dive. We recommend that you first watch the videos and then refer to the documentation as needed for additional explanation. 

Was this helpful?
How can we improve it?