From Bright Pattern Documentation
Jump to: navigation, search
• 日本語

Text To Speech Integration (TTS)

Bright Pattern integrates with the following text-to-speech (TTS) providers for creating voice prompts:

When text-to-speech functionality is enabled, call prompts may use a variety of pre-recorded voices, accents, and languages to automate responses.

General TTS Properties

After you add a TTS account, several configurable properties will appear in the Properties pane. See this section for properties shared by all TTS accounts, or see below for provider-specific properties.

Name

The unique name of this integration account, used to identify it elsewhere in your contact center.

URL

Your Text to Speech provider API URL. For default URLs, see the various TTS sections below.

Load Voices

Allows you to choose a voice type from a preconfigured list of voice models. After clicking Load Voices, wait for confirmation that the voices have loaded before clicking Apply. This may take a few minutes. If the field still says "empty", wait a few more minutes and try again.

Test Connection

Tests the connection between Bright Pattern and the TTS provider. When testing the connection, if the account has been configured correctly, you will see the message "Account credentials appear to be valid."

AWS

This integration uses the Amazon Polly API.

AWS TTS integration properties

URL

https://polly.{region}.amazonaws.com

Region

The geographic region in which your AWS account exists. You may choose a region from the drop-down menu or type one in. It must be set in the format shown in the drop-down menu (i.e., "us-west-2" and not "US West (Oregon)").

Access Key ID

An access key ID associated with a user in your AWS environment. You may wish to set up a dedicated user with a limited access policy for this purpose, for example:

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Sid": "GenerateSpeechWithPolly",
      "Effect": "Allow",
      "Action": [
        "polly:SynthesizeSpeech",
        "polly:DescribeVoices"
      ],
      "Resource": "*"
    }
  ]
}

You can find the access key ID in the IAM Dashboard > Access Management > Users > Security Credentials tab of the user.

Secret Key

The secret access key corresponding to the inputted access key ID.

Secret keys are only shown when the user is created. If you do not have the secret key, you should generate a new one in the IAM Dashboard > Access Management > Users > Security Credentials tab of the user.

Model/Engine

Select the desired Amazon Polly voice engine: standard, neural, long-form, or generative.

Azure

An existing Azure Speech resource is required to configure this integration.

Azure TTS integration properties

URL

https://{region}.tts.speech.microsoft.com

Region

The region associated with your Speech resource. You can find the region in the Azure portal by selecting the desired Speech resource > Resource Management > Keys and Endpoint.

Token

Your Speech resource key. You can find the resource key in the Azure portal by selecting the desired Speech resource > Resource Management > Keys and Endpoint > KEY 1. Note that resource keys only function within the associated resource region.

Cartesia

This integration uses the Cartesia API.

Cartesia TTS integration properties

URL

https://api.cartesia.ai

API key

Your Cartesia API key.

Model/Engine

Select the desired Cartesia model.

ElevenLabs

This integration uses the ElevenLabs API.

ElevenLabs TTS integration properties

URL

https://api.elevenlabs.io

API key

Your ElevenLabs API key.

Model/Engine

Select the desired ElevenLabs model.

Google TTS

Please note: Google Cloud TTS requires a payment method to be attached to the account; without this, the TTS service would silently not work, and there would be an error in the Google API dashboard.

If you see the message An error occurred: HTTP/1.1 403 Forbidden after clicking Load Voices, check that a valid payment method is attached to your Cloud Billing account.


Google Cloud TTS integration properties

URL

https://www.googleapis.com/auth/cloud-platform

API Key

The API key that you created for your Google Cloud APIs and services. See Generate a Vertex AI Service Account Key for more information.

Note that this is not the same as the service key ID.

Google Streaming TTS

This integration uses the Google Streaming API, which reduces the delay before voice playback begins, as compared to non-streaming integrations. This can create more natural interactions when using real-time AI features like AI Agent.

Note that Google only allows Chirp 3: HD voices with this integration. Before clicking Load Voices, please verify that the API key file is correct by clicking the Test Connection button

If you see the message An error occurred: HTTP/1.1 403 Forbidden after clicking Load Voices, check that a valid payment method is attached to your Cloud Billing account.


Google Streaming TTS integration properties

URL

https://www.googleapis.com/auth/cloud-platform

API Key (JSON File)

The API key that you created for your Google Cloud APIs and services. See Generate a Vertex AI Service Account Key for more information.

Rime

This integration uses Rime’s text-to-speech API.

Rime TTS integration properties

URL

https://users.rime.ai

API key

Your Rime API token.

Model/Engine

Select the desired Rime model.

Watson

This integration uses the IBM Text to Speech API cloud service.

Watson TTS integration properties

URL

Your unique IBM Text to Speech service instance URL in the following format:

https://api.{location}.text-to-speech.watson.cloud.ibm.com/instances/{instance id}

For example:

https://api.us-south.text-to-speech.watson.cloud.ibm.com/instances/0123a4c0-5d67-8cec-9c10-a0cdbb1234b5

You can find the URL in your Text to Speech service credentials. You must get the URL from your service credentials on IBM Cloud.

API key

The API key (string) for your IBM Text to Speech service instance, which is used for making secure, authenticated requests to IBM Cloud APIs. You can find the API key in your Text to Speech service credentials.

< Previous | Next >