A is incorrect: Translation converts spoken or written content from one language to another, not for speaker identification.

B is incorrect: Text-to-Speech synthesizes spoken audio from written text, which is the reverse of transcription and not related to speaker identification.

C is correct: Azure AI Speech's Batch Transcription service includes a diarization feature that can distinguish between multiple speakers in an audio file, labeling who spoke when.

D is incorrect: Voice Cloning involves creating a synthetic voice that sounds like a specific person, which is distinct from identifying multiple speakers in an existing recording.