Creating an override file for Watson Text to Speech

An override file enables you to specify how Watson Text to Speech is installed on your environment. You can customize the installation by modifying the YAML file speech-override.yaml. You then specify the file as an installation parameter.

1. Create a speech-override.yaml file and place it in the same directory as the IBM® Cloud Pak for Data command-line interface. You specify the file as an installation parameter. In the file, define any custom installation and configuration settings. A sample file that provides the minimum default settings follows.

   tags:
     sttAsync: true
     sttCustomization: true
     ttsCustomization: true
     sttRuntime: true
     ttsRuntime: true
   
   affinity: {}
   
   global:
     dockerRegistryPrefix: "{Registry}"
     image:
       pullSecret: "{Registry_pull_secret}"
   
     datastores:
       minio:
         serviceAccountName: "ibm-minio-operator"
         #Sizing
         deploymentType: Development #Development or Production
         #Storage
         storageClassName: "portworx-shared-gp3"
         #Secrets
         authSecretName: "minio"
       rabbitMQ:
         #Storage
         storageClassName: "portworx-shared-gp3"
       postgressql:
         #Storage
         databaseStorageClass: "portworx-shared-gp3"
         databaseArchiveStorageClass: "portworx-shared-gp3"
         databaseWalStorageClass: "portworx-shared-gp3"
         #Secrets
         authSecretName: "user-provided-postgressql"
   
     sttModels:
       enUsBroadbandModel:
         enabled: true
         catalogName: en-US_BroadbandModel
       enUsNarrowbandModel:
         enabled: true
         catalogName: en-US_NarrowbandModel
       enUsShortFormNarrowbandModel:
         enabled: true
         catalogName: en-US_ShortForm_NarrowbandModel
   
     ttsVoices:
       enUSMichaelV3Voice:
         enabled: true
         catalogName: en-US_MichaelV3Voice
       enUSLisaV3Voice:
         enabled: true
         catalogName: en-US_LisaV3Voice
       enUSAllisonV3Voice:
         enabled: true
         catalogName: en-US_AllisonV3Voice

   where

   • {Registry} is the path for the internal Docker registry. It must be image-registry.openshift-image-registry.svc:5000/{namespace}, where {namespace} is the namespace in which IBM Cloud Pak for Data is installed, normally zen.

   • {Registry_pull_secret} is the pull secret for the internal Docker registry, which you can learn by running the following command:

     oc get secrets | grep default-dockercfg

   • The authentication secrets for the Minio and PostgresSQL datastores must be created before the installation and must have the following values:

     • For Minio, the authsecretname must be minio.

     • For PostgresSQL, the authsecretname must be user-provided-postgressql.

2. What to do next: Installing Watson Text to Speech service