Oracle Cloud Speech Engine

  • Provider - Cloud service offering the speech engine.
  • OCI User ID - A unique Oracle Cloud Identifier for your account within the Oracle Cloud Infrastructure tenancy. Identifies the user making the API request to Oracle speech.
  • OCI Tenant OCID - A unique Oracle Cloud Identifier for your entire Oracle Cloud Infrastructure tenancy. Identifies the tenancy that owns and manages the Speech resources.
  • OCI Compartment OCID - The Oracle Cloud Identifier that uniquely identifies the resource compartment in the Oracle Cloud Infrastructure.
  • OCI Key Fingerprint - The unique identifier for the API signing or SSH key you created. Used to tell Oracle Cloud which API key you are using to sign (authenticate) your requests. Enter the unique identifier for the API signing or SSH key you created. Used to tell Oracle Cloud which API key you are using to sign (authenticate) your requests.
  • OCI Private Key - The Oracle Cloud private key you created for secure authentication. Used to digitally sign the API requests from Oracle Intelligent Communications Orchestration Network Cloud Service, proving the authenticity of the request coming from an authorized user.
  • OCI Region - The region where the Oracle Cloud Infrastructure speech engine is deployed. For example: us-phoenix-1.

    For a list of valid region codes, see Region and Availability of Domains.

  • Language Code - The BPC-47 language code for speech recognition. For example: en-US. The BCP 47 code specifies the language you want the service to use for speech recognition.

    For a list of valid language codes, see Language and Feature Support.

  • Model Type - Select either the Oracle or Whisper speech recognition model. Refers to the broad category of Machine Learning (ML) model to use for voice transcription. Specify this during speech-to-text configuration.
  • Model Domain - Select either Generic or Medical. Specifies the context in which the speech model is optimized to perform. Specify this during speech-to-text configuration.
  • Model Name - Select either TTS-1 STANDARD or TTS-2 NATURAL. Model Name specifies the voice that you want Oracle Speech to use when synthesizing speech. Specify this during text-to-speech configuration.

    Note:

    The model name directly affects which "voicename" you can set in the Voice Gateway Markup Language in the Oracle Digital Assistant.