Turn audio into searchable, speaker-labeled transcripts. Built for iterative pipelines, not one-shot runs.
