AudioCodes Voice.AI User manual
Product Description
AudioCodes Intuitive Human Communications for Chatbot Services
Voice.AI Gateway
Version 1.6
Product Description Contents
Table of Contents
1 Introduction ......................................................................................................... 7
2 Conversation Flow .............................................................................................. 9
2.1 Conversation Setup ................................................................................................ 9
2.2 Speech Flow......................................................................................................... 10
3 Main Components of Voice.AI Gateway .......................................................... 11
3.1 Voice Engagement Channel Component .............................................................. 11
3.2 Cognitive Services Component............................................................................. 12
3.2.1 Text-to-Speech Services .........................................................................................13
3.2.2 Speech-to-Text Services .........................................................................................13
3.2.3 Bot Frameworks .......................................................................................................14
4 Deployment Options ......................................................................................... 15
4.1 AudioCodes Cloud Account .................................................................................. 15
4.2 Customer's Data Center ....................................................................................... 16
4.2.1 Required Resources ................................................................................................17
5 Performance Monitoring ................................................................................... 19
5.1 CDRs and Call Transcripts ................................................................................... 19
5.2 Fault Management with Alarms ............................................................................ 20
6 Voice.AI Gateway Features .............................................................................. 21
6.1 Bot Frameworks ................................................................................................... 21
6.2 Conversation Initiation Features ........................................................................... 21
6.2.1 Initial Activity to Bot ..................................................................................................21
6.2.1.1 Passing SIP Headers to Bot .....................................................................21
6.2.2 Connect on Bot Prompt ...........................................................................................22
6.2.3 Initial Message to Bot ..............................................................................................22
6.2.4 Welcome Message ..................................................................................................22
6.3 Speech Features .................................................................................................. 22
6.3.1 STT and TTS Providers ...........................................................................................22
6.3.2 Language .................................................................................................................22
6.3.3 Custom Language and STT Context .......................................................................22
6.3.4 SSML for TTS ..........................................................................................................23
6.3.5 Continuous ASR ......................................................................................................23
6.3.6 Overriding STT Parameters for Activation and Streaming ......................................23
6.3.7 Stored STT Transcriptions .......................................................................................23
6.3.8 Punctuation of STT Transcriptions ..........................................................................24
6.3.9 Barge-In by Speech or DTMF Input .........................................................................24
6.3.10 TTS Caching ............................................................................................................24
6.4 Sending User DTMF Digits to Bot ......................................................................... 25
6.5 Activities upon Failure During Call ........................................................................ 25
6.6 Timeouts for User Input and Bot/STT/TTS Responses ......................................... 25
6.7 Playing Prompts to User for Error Handling .......................................................... 25
6.8
Azure Key Vault for Secure Secrets Storage ........................................................ 25
6.9 Call Control .......................................................................................................... 26
6.9.1 Call Transfer ............................................................................................................26
6.9.1.1 Adding SIP Headers on Call Transfer ......................................................26
6.9.2 Disconnect ...............................................................................................................26
Version 1.6 3 Audio-based Chatbot Solutions
Voice.AI Gateway
This page is intentionally left blank.
Product Description 4 Document #: LTRT-30905
Product Description Notices
Notice
Information contained in this document is believed to be accurate and reliable at the time of
printing. However, due to ongoing product improvements and revisions, AudioCodes cannot
guarantee accuracy of printed material after the Date Published nor can it accept responsibility
for errors or omissions. Updates to this document can be downloaded from
https://www.audiocodes.com/library/technical-documents.
This document is subject to change without notice.
Date Published: June-08-2020
WEEE EU Directive
Pursuant to the WEEE EU Directive, electronic and electrical waste must not be disposed of
with unsorted waste. Please contact your local recycling authority for disposal of this product.
Customer Support
Customer technical support and services are provided by AudioCodes or by an authorized
AudioCodes Service Partner. For more information on how to buy technical support for
AudioCodes products and for contact information, please visit our website at
https://www.audiocodes.com/services-support/maintenance-and-support.
Stay in the Loop with AudioCodes
Abbreviations and Terminology
Each abbreviation, unless widely used, is spelled out in full when first used.
Related Documentation
Document Name
Voice.AI Gateway API Reference Guide
Voice.AI Gateway Integration Guide
Voice.AI Gateway with One-Click Dialogflow Integration Guide
Version 1.6 5 Audio-based Chatbot Solutions
Voice.AI Gateway
General Notes, Warnings, and Safety Information
Note: OPEN SOURCE SOFTWARE. Portions of the software may be open source
software and may be governed by and distributed under open source licenses, such
as the terms of the GNU General Public License (GPL), the terms of the Lesser
General Public License (LGPL), BSD and LDAP, which terms are located at
https://www.audiocodes.com/services-support/open-source/ and all are incorporated
herein by reference. If any open source software is provided in object code, and its
accompanying license requires that it be provided in source code as well, Buyer may
receive such source code by contacting AudioCodes, by following the instructions
available on AudioCodes website.
Document Revision Record
LTRT Description
30902 Initial document release.
30903 Features updated; deployment options updated.
30904 Support for Google Dialogflow, Google Cloud Text-to-Speech and Speech-to-Text.
Sections updated: Connect on Bot Prompt (default); Language (voice name); SSML
for TTS; Punctuation of STT Transcriptions (default); Barge-In (seconds); TTS
Caching (default); Call Transfer; Disconnect
30905 Updated to Ver. 1.6
Nuance STT/TTS for trial; fault management using alarms; Azure Key Vault; storing
STT transcriptions; playing prompts to user upon errors; barge-in by DTMF; sending
DTMF; activities upon call failure; overriding STT parameters; timeouts for user input
and bot/STT/TTS response
Documentation Feedback
AudioCodes continually strives to produce high quality documentation. If you have any
comments (suggestions or errors) regarding this document, please fill out the Documentation
Feedback form on our website at https://online.audiocodes.com/documentation-feedback.
Product Description 6 Document #: LTRT-30905
Product Description 1. Introduction
1 Introduction
AudioCodes Voice.AI Gateway enhances chatbot functionality by allowing human
communication with chatbots through voice (voicebot), offering an audio-centric user
experience. Employing AudioCodes' Voice.AI Gateway service provides you with a singlevendor solution, assisting you in migrating your text-based chatbot experience into a voicebased chatbot.
The Voice.AI Gateway also allows you to provide this enhanced chatbot service to a wider
audience that uses various voice engagement channels from where users initiate calls with
chatbots:
SIP Trunk
PSTN or cellular
Enterprise Unified Communications (UC) and IP-PBXs
Contact Centers
WebRTC
AudioCodes’ field-proven and sophisticated voice communications technology embedded in
the Voice.AI Gateway allows seamless integration into any existing voice network. The
Voice.AI Gateway can connect to and integrate with any third-party cognitive service – bot
frameworks, speech-to-text (STT) engines, and text-to-speech (TTS) engines. It can also
operate with TTS and STT engines for multiple languages.
Figure 1-1: Voice Engagement Channel and Cognitive Services of Voice.AI Gateway
Version 1.6 7 Audio-based Chatbot Solutions
Voice.AI Gateway
The Voice.AI Gateway provides advanced call management capabilities, for example, call
disconnect, call transfer to a human agent, and call recording.
As such, the Voice.AI Gateway provides and uses various APIs for different purposes:
SIP, RTP and WebRTC APIs for communicating with voice engagement channels
HTTP-based APIs for interfacing with third-party voice cognitive services that convert
voice to text (STT) and text to voice (TTS)
HTTP-based APIs for connecting to third-party bot frameworks such as Azure,
Google, and AW S
Product Description 8 Document #: LTRT-30905
Product Description 2. Conversation Flow
2 Conversation Flow
This section describes the call flow between the voice engagement channel (telephony) and
the cognitive chatbot services, through AudioCodes Voice.AI Gateway.
2.1 Conversation Setup
The call setup flow between the voice engagement channel (telephony) and the cognitive
chatbot services through AudioCodes Voice.AI Gateway is shown and explained below:
Figure 2-1: Dialog Flow Between Voice Channel and Chatbot Cognitive Services
1. The user initiates a call with the chatbot, sending a SIP INVITE message to the Voice.AI
Gateway.
2. The Voice.AI Gateway processes the call by employing its inherent session border
controller (SBC) functionality. This includes basic call handling stages (call
classification, message manipulation and routing) and optionally, other feature-rich SBC
capabilities. The device searches its IP-to-IP Routing table for a matching routing rule
(typically based on dialed number), which determines the route to a specific bot
framework (destination).
3. The device initiates an HTTP-based dialog setup with a specific bot framework,
according to the destination of the matched routing rule.
4. An HTTP-based connection is established between the Voice.AI Gateway and the bot
framework.
5. The Voice.AI Gateway sends a SIP 200 OK response to the user, establishing the media
session between it and the user.
Version 1.6 9 Audio-based Chatbot Solutions
Loading...