Text-to-Speech (TTS) in Webex Contact Center

Overview

Webex Contact Center offers in-built Text-to-Speech (TTS) capability for customers. TTS automatically converts any written text into speech or synthesized audio to provide Interactive Voice Response (IVR) service to customers. TTS service includes a predefined set of neural voices and language options that you can configure in the IVR application to build call flows.

Note:

By default, Cisco TTS is available for the Flex 3 customers who are onboarded on the Next Generation voice platform.
Existing customers on the Next Generation platform can view both Cisco TTS and Google TTS connectors in the Prompt settings of IVR activities, such as Play Message, Menu, and Collect Digits in the Flow Designer.

Supported languages and voices

Webex Contact Center supports 21 languages and 2 voices per language for Cisco TTS. The following table lists the languages and voices supported for Cisco TTS:

Locale	Language	Male Voice Name	Female Voice Name
de-CH	German (Switzerland)	de-CH-Luca	de-CH-Eva *default
de-DE	German (Germany)	de-DE-Jonas *default	de-DE-Emma
en-GB	English (United Kingdom)	en-GB-Colton	en-GB-Agnes *default
en-IE	English (Ireland)	en-IE-Liam	en-IE-Fiona *default
en-IN	English (India)	en-IN-Veer	en-IN-Ananya *default
en-US	English (United States)	en-US-Daniel	en-US-Maria *default
es-ES	Spanish (Spain)	es-ES-Pablo *default	es-ES-Cristina
es-MX	Spanish (Mexico)	es-MX-Arturo *default	es-MX-Luna
es-US	Spanish (United States)	es-US-Alonso *default	es-US-Paloma
fil-PH	Filipino (Philippines)	fil-PH-Luis	fil-PH-Sofia *default
fr-CA	French (Canada)	fr-CA-Andre *default	fr-CA-Lea
fr-FR	French (France)	fr-FR-Adrien	fr-FR-Ariane *default
it-IT	Italian (Italy)	it-IT-Nico *default	it-IT-Alice
ja-JP	Japanese (Japan)	ja-JP-Haru	ja-JP-Aiko *default
nl-NL	Dutch (Netherlands)	nl-NL-Edwin	nl-NL-Adriana *default
pl-PL	Polish (Poland)	pl-PL-Filip	pl-PL-Natalia *default
pt-BR	Portuguese (Brazil)	pt-BR-Davi	pt-BR-Camila *default
sv-SE	Swedish (Sweden)	sv-SE-Hugo *default	sv-SE-Kristina
th-TH	Thai (Thailand)	th-TH-Chaisai *default	th-TH-Anong
vi-VN	Vietnamese (Vietnam)	vi-VN-Tai	vi-VN-Linh *default
zh-CN	Chinese (Mandarin, Simplified)	zh-CN-Hong	zh-CN-Liu *default

To use the default language and voice for Cisco Text-to-Speech, you must configure the global variables Global_VoiceName and Global_Language in the flow. For more information about how to add these global variables to the flow, see Global Variables.
To override the default language and voice for Cisco Text-to-Speech, configure the Set Variable activity with the appropriate voice name code.

Supported Speech Synthesis Markup Language (SSML) tags

The following table lists the standard SSML tags supported for TTS voice.

The <speak> tag is the root element of all SSML text and is mandatory.

SSML Tag	Description	Attribute
<speak>	Identifies the SSML-enhanced text.
<break>	Adds a pause in your message.	strength time
<say-as>	Controls how special types of words are spoken.	interpret-as format detail
<audio>	Embeds short, pre-recorded audio within your message.	src
<p>	Adds a pause between paragraphs.
<s>	Adds a pause between sentences.
<prosody>	Controls the volume, speaking rate, and pitch of the voice.	rate pitch volume
<sub>	Pronounces acronyms and abbreviations	alias