# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand.
overlay: 1.0.0
info:
  title: 'API Evangelist conversational phrasing for Microsoft Azure Custom Speech Model Evaluations: API'
  version: 1.0.0
extends: openapi/microsoft-azure-custom-speech-model-evaluations-api-openapi.yml
actions:
- target: $.info
  update:
    x-apievangelist-phrasing:
      method: generated
      generated: '2026-09-26'
      generator: build-phrasing.py
      label: Generated by API Evangelist
      operations: 8
- target: $.paths['/evaluations/locales'].get
  update:
    x-apievangelist-phrasing:
      intent: List locales supported for speech evaluations
      effect: read
      questions:
      - Which languages can I run custom speech model evaluations in?
      - Is my locale supported for comparing two speech models?
      instructions:
      - text: List the locales supported for speech model evaluations.
      - text: Show which languages evaluations support.
      method: generated
      generated: '2026-09-26'
- target: $.paths['/evaluations'].get
  update:
    x-apievangelist-phrasing:
      intent: List my speech model evaluations
      effect: read
      questions:
      - What speech model evaluations have I run on this subscription?
      - Can I page through my evaluations with skip and top?
      instructions:
      - text: List all my speech model evaluations.
      - text: Show {top} evaluations after skipping {skip}.
        slots:
          top: query.top
          skip: query.skip
      method: generated
      generated: '2026-09-26'
- target: $.paths['/evaluations'].post
  update:
    x-apievangelist-phrasing:
      intent: Compare two speech models on a dataset
      effect: write
      questions:
      - How do I compare the accuracy of two custom speech models on the same test data?
      - What do I need to start a word error rate evaluation between models?
      instructions:
      - text: Create evaluation {displayName} comparing model {model1} with {model2} on dataset {dataset} in {locale}.
        slots:
          displayName: requestBody.displayName
          model1: requestBody.model1
          model2: requestBody.model2
          dataset: requestBody.dataset
          locale: requestBody.locale
      - text: Start an evaluation named {displayName} of {model1} vs {model2} using dataset {dataset} for locale {locale}.
        slots:
          displayName: requestBody.displayName
          model1: requestBody.model1
          model2: requestBody.model2
          dataset: requestBody.dataset
          locale: requestBody.locale
      method: generated
      generated: '2026-09-26'
- target: $.paths['/evaluations/{id}/files'].get
  update:
    x-apievangelist-phrasing:
      intent: List result files of an evaluation
      effect: read
      questions:
      - Where can I download the result files an evaluation produced?
      - How long are evaluation file download links valid?
      instructions:
      - text: List the files of evaluation {id}.
        slots:
          id: path.id
      - text: Get evaluation {id} files with links valid for {sasValidityInSeconds} seconds.
        slots:
          id: path.id
          sasValidityInSeconds: query.sasValidityInSeconds
      method: generated
      generated: '2026-09-26'
- target: $.paths['/evaluations/{id}/files/{fileId}'].get
  update:
    x-apievangelist-phrasing:
      intent: Get one file from an evaluation
      effect: read
      questions:
      - How do I fetch a single result file from an evaluation?
      - Can I get a download link for one evaluation report file?
      instructions:
      - text: Get file {fileId} from evaluation {id}.
        slots:
          fileId: path.fileId
          id: path.id
      - text: Download link for evaluation {id} file {fileId}.
        slots:
          id: path.id
          fileId: path.fileId
      method: generated
      generated: '2026-09-26'
- target: $.paths['/evaluations/{id}'].get
  update:
    x-apievangelist-phrasing:
      intent: Get a speech model evaluation
      effect: read
      questions:
      - What is the status and result of one evaluation I started?
      - Has my speech model comparison finished running?
      instructions:
      - text: Show evaluation {id}.
        slots:
          id: path.id
      - text: Check the status of speech evaluation {id}.
        slots:
          id: path.id
      method: generated
      generated: '2026-09-26'
- target: $.paths['/evaluations/{id}'].delete
  update:
    x-apievangelist-phrasing:
      intent: Delete a speech model evaluation
      effect: destructive
      questions:
      - Can I delete an old speech model evaluation?
      - How do I clean up evaluations I no longer need?
      instructions:
      - text: Delete evaluation {id}.
        slots:
          id: path.id
      - text: Remove speech model evaluation {id} permanently.
        slots:
          id: path.id
      method: generated
      generated: '2026-09-26'
- target: $.paths['/evaluations/{id}'].patch
  update:
    x-apievangelist-phrasing:
      intent: Rename or describe an evaluation
      effect: write
      questions:
      - Can I rename an evaluation after it has been created?
      - Which evaluation details can I still change later?
      instructions:
      - text: Rename evaluation {id} to {displayName}.
        slots:
          id: path.id
          displayName: requestBody.displayName
      - text: Update the description of evaluation {id} to {description}.
        slots:
          id: path.id
          description: requestBody.description
      method: generated
      generated: '2026-09-26'