A clerk’s office records city council meetings and needs searchable text captions generated from the audio files. Which AWS service converts speech to text for that use case?

A clerk's office that records city council meetings and needs searchable text captions from audio should use Amazon Transcribe for automatic speech-to-text conversion. Transcribe turns spoken audio into text that can be indexed, searched, and published as captions or transcripts—exactly the accessibility and records-search outcome municipal clerks need after long hearings. At Cloud Practitioner depth, identify Transcribe whenever the stem converts speech into text.

Amazon Polly synthesizes spoken audio from written scripts—the opposite direction of the requirement. Amazon Translate converts text from one language into another; it does not create the initial transcript from audio. Amazon Comprehend extracts sentiment and key phrases from finished text; it assumes text already exists. Choosing Polly, Translate, or Comprehend mistakes adjacent AI services for the speech-to-text step.

Exam caveat: Memorize the AI verb map: speech→text = Transcribe; text→speech = Polly; text language A→B = Translate; understand text = Comprehend; see images = Rekognition. CLF thrives on these clean mappings.

Operational check for the clerk's office: send meeting audio (or media pipelines) to Transcribe with an appropriate language settings, store transcripts in S3 with meeting IDs, enable searchable indexes for staff, review accuracy on proper names and ordinances (custom vocabularies help), retain audio plus text per records policy, and publish captions only after a light human review when legal precision matters.