Flowstorm
Product SiteFlowstorm AppAndroid AppiOS App
Search…
What is Flowstorm
Basic principles
Quick Start Guide
Video Tutorials
How To
Sign Up & Login
Design a Dialogue
Create an Application
Analyze Conversations
Tips for Conversation Design
Flowstorm Studio
Working Space
User Settings
Account Management
Keyboard Shortcuts
Flowstorm Assistant
Clients and Channels
Introduction to Clients
Client Configuration
Standalone Application
Android Application
iOS Application
Web client
Amazon Alexa
Google Assistant App
Phone Calls via Twilio.com
Flowstorm Project
About Project
Core Library
Core Application
Channels
NLP Services
Project Setup
Deployment
Boilerplate Project
Development System
Introduction
Dialogue Design
Building Blocks
Dialogue Model
Node Types
Mixins
Snippets
File Assets
Voices
Personas
Applications
Context Scopes
Basic Concepts
Advanced Concepts
DialogueScript
Client Integration
Organization
Platform Continuous Delivery Process
How to publish Core artifacts into Maven Central repository
Powered By GitBook
Voices
​

How Voices work on Amazon Alexa and Google Assistant

If your dialogue is running on Amazon Alexa or Google Assistant then speech synthesis (TTS) is performed on their side instead of in the platform Core runner. As they support only particular voice settings using SSML tag voice we must do some some mapping logic to add this tag into the output and make speech sound as similar as possible to the voice TTS configuration.

Amazon Alexa

a) If applied voice TTS configuration has Amazon Alexa property set, it will result into SSML
1
<voice name="${ttsConfig.amazonAlexa}">$ssml</voice>
Copied!
Set Amazon Alexa property on voices using different TTS provider then Amazon or if Amazon (Amazon Polly service) voice name is not supported by Alexa.
b) If applied voice TTS configuration has Provider set to Amazon value, it will use Name property value for SSML
1
<voice name="${ttsConfig.name}">$ssml</voice>
Copied!
Please ensure that TSS configuration voice name is supported by Amazon Alexa - if not, set appropriate voice name explicitly using Amazon Alexa property (see a) block).
c) otherwise no <voice> tag will be applied to your speech (default Alexa voice will be used)
To see list of voice names supported by Alexa please take a look on documentation page here.

Google Assistant

Applied voice TTS configuration Gender property will be used together with optionally set Google Assistant property value (if it is empty, value 1 will be used instead) used as as assistant voice variant
1
<voice gender="${ttsConfig.gender}" variant="${ttsConfig.googleAssistant ?: 1}">$ssml</voice>
Copied!
Previous
File Assets
Next
Personas
Last modified 1yr ago
Copy link
Contents
How Voices work on Amazon Alexa and Google Assistant
Amazon Alexa
Google Assistant