What was introduced?
We’re saying the provision of AWS Contact Center Intelligence (CCI) solutions, a mix of companies that empowers prospects to simply combine AI into contact facilities, made obtainable by means of AWS Partner Network (APN) companions.
AWS CCI has options for self-service, live-call analytics & agent help, and post-call analytics, making it doable for purchasers to rapidly deploy AI into their present workflows or construct utterly new ones.
What’s AWS Contact Heart Intelligence?
We talked about that AWS CCI brings options to contact facilities powered by AI for earlier than, throughout, and after buyer interactions.
My colleague Swami Sivasubramanian (VP, Amazon Machine Studying, AWS) stated: “We want to make it easy for our customers with contact centers to benefit from machine learning capabilities even if they have no machine learning expertise. By partnering with APN technology and consulting partners to bring AWS Contact Center Intelligence solutions to market, we are making it easier for customers to realize the benefits of cloud-based machine learning services while removing the heavy lifting and the need to hire specialized developers to integrate the ML capabilities in to their existing contact centers.”
However what does that imply? 🤔
AWS CCI options permits you to leverage machine studying performance akin to transcription, text-to-speech, translation, enterprise search, chatbots, enterprise intelligence, and language comprehension into present contact middle environments. Clients can now implement contact middle intelligence ML options to help self-service, live-call analytics & agent help, and post-call analytics. At present, AWS CCI options can be found by means of companions akin to Genesys, Vonage, and UiPath for simple integration into present enterprise contact middle techniques.
“We’re proud Genesys customers will be among the first to benefit from the off-the-shelf machine learning capabilities of AWS Contact Center Intelligence solutions. It’s now simpler and more cost-effective for organizations to combine AWS’s AI capabilities, including search, text-to-speech and natural language understanding, with the advanced contact center capabilities of Genesys Cloud to give customers outstanding self-service experiences.” ~ Olivier Jouve (Government Vice President and Common Supervisor of Genesys Cloud)
“More and more consumers are relying on automated methods to interact with brands, especially in today’s retail environment where online shopping is taking a front seat. The Genesys Cloud and Amazon Web Services (AWS) integration will make it easier to leverage conversational AI so we can provide more effective self-service experiences for our customers.” ~ Aarde Cosseboom (Senior Director of International Member Companies Know-how, Analytics and Product at TechStyle Trend Group)
The way it works and who it’s for…
AWS Contact Heart Intelligence options provide quite a lot of ways in which organizations can rapidly and cost-effectively add machine learning-based intelligence to their contact facilities, by way of AWS pre-trained AI Companies. AWS CCI is at the moment obtainable by means of collaborating APN companions, and it’s centered on three levels of the contact middle workflow: Self-Service, Stay Name Analytics and Agent Assist, and Put up-Name Analytics. Let’s break every one in every of these up.
The Self-Service answer helps with creation of chatbots and ML-driven IVRs (Interactive voice response) to handle the commonest queries a contact middle workforce usually will get. This now permits precise name middle staff to concentrate on greater worth work. To implement this answer, you’ll need to work with both Amazon Lex and/or Amazon Kendra. The novelty of this answer is that Lex + Kendra not solely fulfills transactional queries (i.e. ebook a lodge room or reset my password), but in addition addresses the lengthy tail of shoppers questions whose solutions reside in enterprises data techniques. Earlier than, these Q&A needed to be arduous coded in Amazon Lex, making it tougher to implement and preserve. Right now, you may implement this answer instantly out of your present contact middle platform with AWS CCI companions, akin to Genesys.
The Stay Name Analytics & Agent Help answer permits the creation of real-time ML capabilities to extend workers productiveness and engagement. Right here, Amazon Transcribe is used to carry out real-time speech transcription, whereas Amazon Comprehend can analyze interactions, detect the sentiment of the caller, and determine key phrases and phrases within the dialog. Amazon Translate may even be added to translate the dialog right into a most well-liked language! Now, you may implement this answer instantly from a number of main contact middle platforms with AWS CCI companions, like SuccessKPI.
The Put up-Name Analytics answer is an automated evaluation of contact middle conversations, which have a tendency to depart actionable knowledge for product and repair suggestions loops. Just like reside name analytics, this answer combines Amazon Transcribe to carry out speech recognition and creates a high-quality textual content transcription of every name, with Amazon Comprehend to investigate the interplay. Amazon Translate might be added to translate the dialog into your most well-liked language, and Amazon Kendra can be utilized for contextual pure language queries. Right now, you may implement this answer instantly from a number of main contact middle platforms with AWS CCI companions, akin to Acqueon.
AWS helps companions combine these options into their merchandise. Some options even have a Quick Start, which incorporates CloudFormation templates and deployment information, to automate the deployments. The excellent news is that our AWS Companions touchdown pages will even present further implementation data particular to their merchandise. 👌
Let’s see a demo…
For at present’s publish, we selected to concentrate on diving deeper into the Self-Service and Put up-Name Analytics options, so let’s start with Self-Service.
Now we have a public GitHub repository that has a complete Quick Start template plus an in depth deployment information with structure diagrams. (And the excellent news is that our APN associate touchdown pages will even reference this repo!)
This GitHub repo talks concerning the Amazon Lex chatbot integration with Amazon Kendra. The principle thought right here is that the client can convey their very own doc repository by means of Amazon Kendra, which might be sourced by means of Amazon Lex when prospects are interacting with this Lex chatbot.
The principle factor we need to discover on this structure is that prospects can convey their present paperwork and permit their chatbot to go looking that doc every time somebody interacts with stated chatbot. The structure beneath assumes our docs are in an S3 bucket, nevertheless it’s value noting that Amazon Kendra can combine with a number of varieties of knowledge sources. If utilizing an S3 bucket, prospects should present their very own S3 bucket title, the one which has their doc repository. It is a prerequisite for deployment.
Let’s observe the directions beneath the repo’s Deployment Steps, skipping forward to Step #2, “Click Deploy to launch the CloudFormation template.”
Since this can be a Fast Begin template, you may see how the whole lot is already stuffed out for us. We click on Subsequent and transfer on to Step 2, Specify stack particulars.
Discover how the S3 bucket part is clean. You’ll be able to present your individual S3 bucket title if you wish to check this out with your individual docs. For at present, I’m going to make use of the S3 bucket title that was supplied to us within the GitHub doc.
The subsequent half to configure would be the Cross account position configuration part. For my demo, I’ll add my very own AWS account ID beneath “Assuming Account ID.”
We click on Subsequent and transfer on to Step 3, Configure Stack choices.
Nothing to configure right here, so we will click on Subsequent once more and transfer on to Step 4, Evaluation. We click on to simply accept these ultimate acknowledgements and click on Create Stack.
If we have been to navigate over to our deployed AWS CloudFormation stacks, we will go to Outputs of this stack and see our Kendra index title and Lex bot title.
Now if we head over to Amazon Lex, we should always be capable to simply discover our chatbot.
We click on into it and we will see that our chatbot is prepared. At this level, we will begin interacting with it!
We will one thing like “Hi” for instance.
Finally we’d additionally get a response that particulars the reply supply. What this implies is that it’ll let you know if this got here from Amazon Lex or from Amazon Kendra and the paperwork we saved in our S3 bucket.
Stay Name Analytics & Agent Help
Now we have two public GitHub repositories for this answer too, and each have detailed deployment information with structure diagrams as properly.
This GitHub repo gives us a code instance and a totally useful AWS Lambda perform to get you began with capturing and transcribing Amazon Chime Voice Connector telephone calls utilizing Amazon Kinesis Video Streams and Amazon Transcribe. This answer provides us the flexibility to see the right way to use AI and ML companies to speak to the client’s existent atmosphere, to drive agent help or analytics. We will take a real-time voice feed, transcribe that data, after which use Amazon Comprehend to drag that data out to supply the important thing motion and sentiment.
We now additionally present the Chime SIP req connector (a chime part that means that you can join voice over an IP appropriate atmosphere with Amazon voice companies) to stream voice in Amazon Transcribe from nearly any contact middle. Our associate Vonage can do the identical by means of websocket.
👉🏽 Take a look at the GitHub developer docs:
And as we talked about above, for at present’s publish, we selected to concentrate on diving deeper into the Self-Service and Put up-Name Analytics options. So let’s transfer on to indicate an instance for Put up-Name Analytics.
Put up-Name Analytics
Now we have a public GitHub repository for this solution too, with one other full Fast Begin template and detailed deployment information with structure diagrams. This answer is used after the decision has ended, in order that our prospects can evaluate the analytics of these calls.
This GitHub repo talks about the right way to search for insights and details about calls which have already occurred. We name this, High quality Administration. We will use Amazon Transcribe and Amazon Comprehend to drag out key phrases, data, and knowledge, to be able to know the right way to higher drive what is occurring in our contact middle calls. We will then evaluate these insights on Amazon QuickSight.
Let’s take a look at the structure diagram for this answer too. Our name recording will get saved in an S3 bucket, which is then picked up by a Lambda perform which does a transcription utilizing Amazon Transcribe. It places the lead to a unique bucket after which that decision’s metadata will get saved in DynamoDB. Now Amazon Comprehend can conduct textual content evaluation on the decision’s metadata, and shops the lead to a Textual content evaluation Output bucket. Finally, QuickSight is used to supply dashboards exhibiting the ensuing name analytics.
Identical to within the earlier instance, we transfer all the way down to the Deployment steps part. Identical to earlier than, we’ve a pre-made CloudFormation template that is able to be deployed.
Step 1, Specify template is nice to go, so we click on Subsequent.
In Step 2, Specify stack particulars, one thing essential to notice is that the Person Pool Area Identify have to be globally distinctive.
We click on Subsequent and transfer on to Step 3, Configure Stack choices. Nothing further to configure right here both, so we will click on Subsequent once more and transfer on to Step 4, Evaluation.
We click on to simply accept these ultimate acknowledgements and click on Create Stack.
And if we have been to navigate over to our deployed AWS CloudFormation stacks once more, we will go to Outputs of this stack and see the PortalEndpoint key. After the stack creation has accomplished efficiently, and portal web site is offered at CloudFront distribution endpoint. This key’s what’s going to permit us to search out the portal URL.
We might want to have consumer created in Amazon Cognito for the subsequent steps to work. (When you have by no means created one, go to this how-to guide.)
⚠️ NOTE: Be sure to open the portal URL endpoint in a unique Incognito Window because the portal attaches a QuickSight Person Position that may intrude along with your precise position.
We go to the portal URL and login with our created Cognito consumer. We’re prompted to alter the non permanent password and are ultimately directed to the QuickSight homepage.
Now we need to add the audio information of our calls and we will achieve this with the Add button.
After efficiently importing our audio information, the audio processing will run by means of transcription and textual content evaluation. At this level we will click on on the Name Analytics brand within the prime left of the Navigation Bar to return to dwelling web page.
Now we will drill down right into a name to see Amazon Comprehend’s results of the decision classifications and turn-by-turn sentiments.
We’re saying AWS CCI availability with 12 APN companions: Genesys, UiPath, Vonage, Acqueon, SuccessKPI, and Inference Solutions (Know-how companions), and Slalom, Onica/Rackspace, TensorIoT, Quantiphi, Accenture, and HGS Digital (Consulting companions).
Able to get began? Contact one of many AWS CCI launch companions listed on the AWS CCI web page.
You might also need to see…
👉🏽AWS Fast Begin hyperlinks from publish:
¡Gracias por tu tiempo!
~Alejandra 💁🏻♀️🤖 y Canela 🐾