Question

A media company needs to automatically generate subtitles in 12 languages for a large library of pre-recorded training videos. They want a low-code solution that transcribes the spoken audio and then translates it, minimizing custom model development. Which combination of Google Cloud AI APIs should the ML engineer recommend?

Accepted Answer

Use the Speech-to-Text API to transcribe the audio, then pass the transcripts to the Cloud Translation API for the 12 target languages.

Answer

Train a custom AutoML translation model for each of the 12 languages using the video transcripts as training data.

Answer

Use the Video Intelligence API label detection feature, then translate the detected labels with the Cloud Translation API.

Answer

Use Document AI to extract text from video frames, then apply the Natural Language API for language detection and translation.