# 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 Model API
  version: 1.0.0
extends: openapi/microsoft-azure-model-api-openapi.yml
actions:
- target: $.info
  update:
    x-apievangelist-phrasing:
      method: generated
      generated: '2026-10-01'
      generator: build-phrasing.py
      label: Generated by API Evangelist
      operations: 11
- target: $.paths['/modelmanagement/v1.0/subscriptions/{subscriptionId}/resourceGroups/{resourceGroup}/providers/Microsoft.MachineLearningServices/workspaces/{workspace}/models/{id}'].get
  update:
    x-apievangelist-phrasing:
      intent: Get a registered machine learning model
      effect: read
      questions:
      - What framework and version is a registered model in my Azure Machine Learning workspace?
      - Can I look up a registered ML model by its id?
      instructions:
      - text: Get registered model {id} from workspace {workspace} in resource group {resourceGroup}, subscription {subscriptionId}.
        slots:
          id: path.id
          workspace: path.workspace
          resourceGroup: path.resourceGroup
          subscriptionId: path.subscriptionId
      - text: Show the details of ML model {id} in workspace {workspace}.
        slots:
          id: path.id
          workspace: path.workspace
      method: generated
      generated: '2026-09-26'
- target: $.paths['/modelmanagement/v1.0/subscriptions/{subscriptionId}/resourceGroups/{resourceGroup}/providers/Microsoft.MachineLearningServices/workspaces/{workspace}/models/{id}'].delete
  update:
    x-apievangelist-phrasing:
      intent: Delete a registered machine learning model
      effect: destructive
      questions:
      - How do I remove a model from my Azure Machine Learning model registry?
      - Does deleting a registered model fail if it has already been removed?
      instructions:
      - text: Delete registered model {id} from workspace {workspace} in resource group {resourceGroup}, subscription {subscriptionId}.
        slots:
          id: path.id
          workspace: path.workspace
          resourceGroup: path.resourceGroup
          subscriptionId: path.subscriptionId
      - text: Unregister ML model {id} in workspace {workspace}.
        slots:
          id: path.id
          workspace: path.workspace
      method: generated
      generated: '2026-09-26'
- target: $.paths['/modelmanagement/v1.0/subscriptions/{subscriptionId}/resourceGroups/{resourceGroup}/providers/Microsoft.MachineLearningServices/workspaces/{workspace}/models/{id}'].patch
  update:
    x-apievangelist-phrasing:
      intent: Patch a registered machine learning model
      effect: write
      questions:
      - Can I change the description or tags of a model I already registered?
      - How do I apply a patch to an existing registered ML model?
      instructions:
      - text: Patch registered model {id} in workspace {workspace}, resource group {resourceGroup}, subscription {subscriptionId}.
        slots:
          id: path.id
          workspace: path.workspace
          resourceGroup: path.resourceGroup
          subscriptionId: path.subscriptionId
      - text: Apply my changes to the existing ML model {id} in workspace {workspace}.
        slots:
          id: path.id
          workspace: path.workspace
      method: generated
      generated: '2026-09-26'
- target: $.paths['/modelmanagement/v1.0/subscriptions/{subscriptionId}/resourceGroups/{resourceGroup}/providers/Microsoft.MachineLearningServices/workspaces/{workspace}/models'].get
  update:
    x-apievangelist-phrasing:
      intent: Search registered models in an ML workspace
      effect: read
      questions:
      - Which models are registered in my Azure Machine Learning workspace?
      - Can I find registered models by name, framework or tag?
      - How do I list only the models produced by a particular training run?
      instructions:
      - text: List registered models named {name} in workspace {workspace}, resource group {resourceGroup}, subscription {subscriptionId}.
        slots:
          name: query.name
          workspace: path.workspace
          resourceGroup: path.resourceGroup
          subscriptionId: path.subscriptionId
      - text: Find models built with framework {framework} in workspace {workspace}.
        slots:
          framework: query.framework
          workspace: path.workspace
      - text: List models from run {runId} in workspace {workspace}.
        slots:
          runId: query.runId
          workspace: path.workspace
      method: generated
      generated: '2026-09-26'
- target: $.paths['/modelmanagement/v1.0/subscriptions/{subscriptionId}/resourceGroups/{resourceGroup}/providers/Microsoft.MachineLearningServices/workspaces/{workspace}/models'].post
  update:
    x-apievangelist-phrasing:
      intent: Register a machine learning model
      effect: write
      questions:
      - How do I register a trained model file in my Azure Machine Learning workspace?
      - What do I need to supply besides the model URL and MIME type when registering?
      instructions:
      - text: Register model {name} from {url} with MIME type {mimeType} in workspace {workspace}, resource group {resourceGroup}, subscription {subscriptionId}.
        slots:
          name: requestBody.name
          url: requestBody.url
          mimeType: requestBody.mimeType
          workspace: path.workspace
          resourceGroup: path.resourceGroup
          subscriptionId: path.subscriptionId
      - text: Register {name} ({mimeType}) at {url} in workspace {workspace} as framework {framework}.
        slots:
          name: requestBody.name
          mimeType: requestBody.mimeType
          url: requestBody.url
          workspace: path.workspace
          framework: requestBody.framework
      method: generated
      generated: '2026-09-26'
- target: $.paths['/modelmanagement/v1.0/subscriptions/{subscriptionId}/resourceGroups/{resourceGroup}/providers/Microsoft.MachineLearningServices/workspaces/{workspace}/models/{id}/metrics'].get
  update:
    x-apievangelist-phrasing:
      intent: Get operational metrics for a registered model
      effect: read
      questions:
      - What operational events have been collected for one of my registered models?
      - Can I pull model metrics for a specific date range?
      instructions:
      - text: Get metrics for model {id} in workspace {workspace}, resource group {resourceGroup}, subscription {subscriptionId}.
        slots:
          id: path.id
          workspace: path.workspace
          resourceGroup: path.resourceGroup
          subscriptionId: path.subscriptionId
      - text: Show operational events for ML model {id} in {workspace} between {startDate} and {endDate}.
        slots:
          id: path.id
          workspace: path.workspace
          startDate: query.startDate
          endDate: query.endDate
      method: generated
      generated: '2026-09-26'
- target: $.paths['/modelrecipes'].get
  update:
    x-apievangelist-phrasing:
      intent: List recipes for building custom voice models
      effect: read
      questions:
      - Which recipes can I use to train a custom neural voice model?
      - Does any voice model recipe support multi-style voices?
      instructions:
      - text: List the supported custom voice model recipes for API version {api_version}.
        slots:
          api_version: query.api-version
      - text: Show which voice training recipes are available.
      method: generated
      generated: '2026-09-26'
- target: $.paths['/models'].get
  update:
    x-apievangelist-phrasing:
      intent: List custom voice models
      effect: read
      questions:
      - Which custom voice models exist on my Speech resource?
      - Can I filter my voice models by project or locale and page through them?
      instructions:
      - text: List the custom voice models on my Speech resource.
      - text: List voice models matching {filter}, skipping the first {skip}.
        slots:
          filter: query.filter
          skip: query.skip
      method: generated
      generated: '2026-10-01'
- target: $.paths['/models/{id}'].get
  update:
    x-apievangelist-phrasing:
      intent: Get a custom voice model
      effect: read
      questions:
      - How do I check whether my custom voice model has finished training?
      - What recipe and training set was a particular voice model built from?
      instructions:
      - text: Get custom voice model {id}.
        slots:
          id: path.id
      - text: Show the training status of voice model {id}.
        slots:
          id: path.id
      method: generated
      generated: '2026-10-01'
- target: $.paths['/models/{id}'].put
  update:
    x-apievangelist-phrasing:
      intent: Train a new custom voice model
      effect: write
      questions:
      - How do I train a custom neural voice from my training set and talent consent?
      - Which recipe, project and consent do I need to build a voice model?
      instructions:
      - text: Create voice model {id} in project {project} from training set {trainingSet} with consent {consent} using recipe {recipe}.
        slots:
          id: path.id
          project: requestBody.projectId
          trainingSet: requestBody.trainingSetId
          consent: requestBody.consentId
          recipe: requestBody.recipe
      - text: Train custom voice {voiceName} as model {id} for locale {locale}.
        slots:
          voiceName: requestBody.voiceName
          id: path.id
          locale: requestBody.locale
      method: generated
      generated: '2026-10-01'
- target: $.paths['/models/{id}'].delete
  update:
    x-apievangelist-phrasing:
      intent: Delete a custom voice model
      effect: destructive
      questions:
      - How do I get rid of a custom voice model I no longer need?
      - Can I permanently remove a trained voice model from my Speech resource?
      instructions:
      - text: Delete custom voice model {id}.
        slots:
          id: path.id
      - text: Remove trained voice model {id} from my Speech resource.
        slots:
          id: path.id
      method: generated
      generated: '2026-10-01'