Contact block: Set voice - Amazon Connect

Contact block: Set voice


  • Sets the text-to-speech (TTS) language and voice to use for the contact flow.

  • The default voice is configured to Joanna (Conversational speaking style).

  • You can choose Override speaking style to make it and other voices Amazon Polly Neural Text-to-Speech (NTTS). Neural voices make automated conversations sound more lifelike by improving the pitch, inflection, intonation, and tempo.

    For a list of supported neural voices, see Neural Voices in the Amazon Polly Developer Guide.

  • After this block is run, any TTS invocation resolves to the neural or standard voice selected.

  • If this block is triggered during a chat conversation, the contact goes down the Success branch. It has no effect on the chat experience.

Supported channels

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

Channel Supported?




No - Success branch


No - Success branch

Contact flow types

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

  • All flows


Configuration tips

  • For the Joanna and Matthew neural voices, in American English (en-US), you can also specify a Conversational speaking style or a Newscaster speaking style.

  • If you're using an Amazon Lex V2 bot, your language attribute in Amazon Connect must match the language model used to build your Lex bot. This is different than Amazon Lex (Classic).

    • If you build an Amazon Lex V2 bot with a different language model—for example, en_AU, fr_FR, es_ES, and more—under Voice, choose a voice that corresponds to that language, and then must choose Set language attribute, as shown in the following image.

    • If you're not using an en-US voice with an Amazon Lex V2 bot and don't choose Set language attribute, the Contact block: Get customer input results in an error.

    • For bots with multiple languages (for example, en_AU and en_GB) choose Set language attribute for one of the languages.

Configured block

When this block is configured, it looks similar to the following image:


See these topics for scenarios that use this block: