Contact block: Play prompt - Amazon Connect

Contact block: Play prompt

Description

  • This block can play an interruptible audio prompt, play a text-to-speech message, or send a chat response.

  • Amazon Connect includes a set of pre-recorded prompts for you to use.

  • To use your own voice prompts, you have the following options:

    • Use the Amazon Connect library: You can record and upload your audio prompts in using the Amazon Connect admin console. For instructions, see Create prompts.

    • Use Amazon S3: You can store as many voice prompts as needed in Amazon S3 and access them in real time by using contact attributes in the Play prompt block.

      For example, based on a customer's preferred language, you can dynamically play a voice prompt with a local accent that greets them and thanks them for being a loyal member. You can even concatenate multiple attributes, such as line of business or language preference to create personalized interactions; for example, reservations + Spanish language.

Requirements

  • Amazon Connect supports .wav files to use for your prompt. You must use .wav files that are 8KHz, and mono channel audio with U-Law encoding. Otherwise, the prompt won't play correctly. You can use publicly available third-party tools to convert your .wav files to U-Law encoding. After converting the files, upload them to Amazon Connect.

  • Amazon Connect supports prompts that are less than 50MB and less than five minutes long.

  • When storing prompts in an S3 bucket: for Regions that are disabled by default (also called opt-in Regions) such as US West (Oregon), Africa (Cape Town), and AWS GovCloud (US-West), your bucket must be in the same Region.

Supported channels

The following table lists how this block routes a contact who is using the specified channel.

Channel Supported?

Voice

Yes

Chat

Yes

Task

No - takes the Okay branch but it has no effect

Contact flow types

You can use this block in the following contact flow types:

  • Inbound contact flow

  • Customer Queue flow

  • Customer Whisper flow. You can play prompts from the Amazon Connect library but not prompts stored in Amazon S3.

  • Agent Whisper flow: You can play prompts from the Amazon Connect library but not prompts stored in Amazon S3.

  • Transfer to Agent flow

  • Transfer to Queue flow

Properties

The properties provide you different ways to choose the prompt to be played.

Prompts stored in library

Select from the prompt library (audio): Choose from one of the pre-recorded prompts included with Amazon Connect, or use the Amazon Connect admin console to record and upload your own prompt. There's no way to upload prompts in bulk.

Prompts stored in Amazon S3

Specify an audio file from an S3 bucket: Store as many prompts as you need in an S3 bucket and then refer to them by specifying the bucket path. For best performance, we recommend creating the S3 bucket in the same Region as your Amazon Connect instance.

The following image shows an example of how to set the file path manually.

The following image shows how to specify the S3 bucket path using attributes:

You can provide the S3 path with concatenation, as shown in the following example. This enables you to personalize the prompt, for example, by line of business and language.

Note that only the last part of path is shown in the image but you would enter the full path, for example:

https://example.s3.amazon.aws.com/$['Attributes']['Language']/$['Attributes']['LOB']/1.wav

The following example shows how you can specify the S3 path dynamically using a user-defined attribute.

Text-to-speech or chat text

You can enter plain text, or SSML, as shown in the following image:

SSML-enhanced input text gives you more control over how Amazon Connect generates speech from the text you provide. You can customize and control aspects of speech such as pronunciation, volume, and speed.

For a list of SSML tags you can use with Amazon Connect, see SSML tags supported by Amazon Connect.

For more information, see Add text-to-speech to prompts.

Configuration tips

  • For step-by-step instructions about how to set up a dynamic prompt using contact attributes, see Dynamically select which prompts to play.

  • When playing prompts from an S3 bucket, for best performance we recommend creating the bucket in the same Region as your Amazon Connect instance.

  • When you use text, either for text-to-speech or chat, you can use a maximum of 3,000 billed characters, which is 6,000 characters total. You can also specify text in a flow using a contact attribute.

  • Some existing contact flows have a version of the Play prompt block that doesn't have an Error branch. In this case, the Okay branch will always be taken at runtime. If you update the configuration of a Play prompt block that doesn't have an Error branch, an Error branch will be added to the block automatically in the editor.

  • A contact is routed down the Error branch in the following situations:

    • Amazon Connect is unable to download the prompt from S3. This may be due to an incorrect file path, or the S3 bucket policy is not set up correctly and Amazon Connect does not have access. For instructions about how to apply the policy, and a template you can use, see Set up prompts to play from an S3 bucket.

    • Incorrect audio file format. Only .wav files are supported.

    • The audio file is larger than 50MB or longer than five minutes.

    • The SSML is incorrect.

    • The text-to-speech length exceeds 6000 characters.

    • The the Amazon Resource Name (ARN) for the prompt is incorrect.

Configured block

The following image shows what this block looks like when it's configured for text-to-speech:

The following image shows what this block looks like when it's configured for an S3 bucket:

Sample flows

All of the sample flows use the Play prompt block. Take a look at the Sample inbound flow (first contact experience) to see a Play prompt for chat and one for audio.