Question

A media company receives thousands of hours of user-uploaded video each week. They want to automatically generate text summaries and answer questions about the audio, spoken content, and visual scenes within each video using a single foundation model. Which Google Cloud AI capability best fits this need?

Accepted Answer

Use Gemini's multimodal capabilities to process video, audio, and visual context and produce summaries

Answer

Use the Cloud Vision API to detect objects in individual video frames

Answer

Use the Speech-to-Text API to transcribe the audio track only

Answer

Use BigQuery ML to train a classification model on video metadata