A media company wants to publish English-language interview recordings to international audiences. The plan is to first produce an accurate written transcript of each recording, then convert that transcript into subtitle text in French, German, and Portuguese. Which two AI capabilities does this workflow require? (Select TWO.)

The required capabilities are speech-to-text and text translation. Producing an accurate written transcript from the spoken interview requires speech-to-text, which converts recorded audio into text. Once that English transcript exists, converting it into French, German, and Portuguese subtitle text requires text translation, which converts written content from one language into another while preserving meaning. Together these two capabilities cover the full workflow described: audio becomes English text, and English text becomes subtitle text in three additional languages. Text-to-speech is incorrect because it converts written text into spoken audio, which would be needed if the company wanted dubbed voice tracks in the other languages, but the plan here calls for subtitle text, not spoken audio, so this capability is not part of the workflow. Sentiment analysis is incorrect because it evaluates the emotional tone of text, such as whether a message reads as positive or negative, which plays no role in transcribing spoken words or translating written subtitles into other languages. The workflow as described stops at producing translated subtitle text, so only the transcription and translation capabilities are actually needed.