- Google’s Notebooklm now supports general video description in dozen languages
- The update offers video views the same language options as audio descriptions
- Notebooklm creates your summaries from real loaded documents
Google’s Notebooklm debuted for the first time in its research tool with a general audio description capable of making ‘podcasts’ produced in AI with digital hosts. The logical extension in general video views continued, with a synthetic voice accompanied by a series of slides that include text and illustrations. However, that capacity was only available in English so far.
Notebooklm video descriptions are now available in more than 80 languages. For most people, that translates into translatable versions of video tutorials based on their loaded notes and linked materials. The teacher’s teacher is there and is now a multilingual expert, since they present their own content in everything from Tamil to Polish and beyond.
These are not summaries of the scratch of the web or hallucinated depending on a vague warning. Notebooklm is based on the real material that rises. Everything that the AI, in video or audio says, is extracted directly from its documents, not of generic training data.
Of course, the video in this context is not a cinematographic masterpiece. Slides are not completely animated explanants. This is not Tiktok for term documents; It’s more like PowerPoint for people who don’t want to do PowerPoints. The objective is clarity, not the show.
Global Video of AI
However, that is not the only global update to Notebooklm. While audio descriptions had recently been available in many languages, they were limited to brief outstanding aspects. Now, everyone obtains the full AI audio experience as an alternative to the video option.
For those who want to read a white paper while driving or cooked, this is enormously practical. He will not win a Grammy, but he could help him understand a textbook or a complex report. Professionals who work internationally could use it to summarize a week of fulfilling transcripts such as shared videos or Audio to Portuguese audio summaries. It is not necessary to trust the English understanding of a colleague.
If you want to see how AI can digest and explain its collection of academic documents, blog posts and videos of YouTube, you can produce a narrated video loading its sources as usual, then clicking on the GENERAL DESCRIPTION button. Soon, a video of approximately seven minutes will be ready to share, download or use as you want.
That does not mean that this solves everything. AI can still fight with nuances, for example. But the reliability of the presentations is valuable by itself. Now, they also look good.
