Real-Time Analytics during Live Calls using Amazon Transcribe

Overview

Real-time analytics is the best way for websites to gather user data. This data is used to improve the website and keep everyone safe. Amazon released a real-time analytics tool for websites called Transcribe. It’s an accessible, easy-to-use platform for gathering data from speeches and audio files.

Introduction

Real-time analytics allow for quick feedback and corrections for speech or audio files. It’s especially useful for video files where multiple people can speak simultaneously. These corrections are much faster with real-time analytics versus waiting until the transcription is finished. Additionally, real-time analytics allows organizations to track the most popular words and phrases used by their audience members. This data is crucial in building custom user experiences.

The call transcript also removes the need for customers to repeat themselves if a call is transferred to a different agent. This will help your agents and supervisors better understand customer conversations more accurately. With this information, your supervisors can identify potential customer issues, training opportunities for agents, and call trends more quickly.

It is possible to integrate customer sentiments and agent concerns, such as unspoken time, interruptions, and conversation rate, into a call analytics tool once the data is analyzed. AWS customers can gain valuable insights from their customer’s conversations, like client and agent sentiments and characteristics of conversations such as non-talk time, interruptions, volume, and speaking rate, using Amazon Transcribe Call Analytics. According to Amazon Transcribe Call Analytics, consumers can identify and remove customer information, such as names, from audio and text recordings using the transcription.

Amazon Transcribe Call Analytics also provides real-time transcripts of real-time conversations, which supervisors can use to quickly gain insight into customer interactions and evaluate the appropriate actions. Real-time call analytics offers APIs for developers to transcribe live calls accurately while identifying real-time customer experience issues and sentiment. To help enterprises deliver better customer service and experiences, AWS has updated its Automated Speech Recognition (ASR) service, Amazon Transcribe, to offer real-time call analytics

Adding real-time analytics to our websites to gather analytics data is easy. We only need an Amazon account and a web server with PHP support. First, sign into The Amazon account and access the Transcribe tool via amazonservices-webscapes/transcribe-service/. After that, follow the steps listed in the Amazon documentation for setting up your web server with PHP support. From there, we’re ready to start gathering real-time analytics data!

Exploring Real-Time Call Analytics in the Console

1. Open Amazon Transcribe Service in your AWS account
2. You can see Amazon Transcribe Call Analytics on the left side of the console. Click on Category Management from the navigation pane and then click on Create Category

3. Enter the name of a Category you want, and select REAL_TIME in the Category type dropdown. Then, choose Create from scratch in the Category creation method.

4. We need one rule for this category. Select Transcript content match in the Rule type dropdown. Now, choose to trigger the rule when any words are mentioned during the entire call, and the speaker is either the customer or the agent. Now, enter the words or phrases to look for in the transcript. In this case, enter cancel, canceled, canceled, manager, and supervisor. In your case, you might be more specific depending on your business. For example, if subscriptions are your business, you can look for the phrase cancel my subscription.

Finally, click on create category.

Now that the category has been created. We have selected Real-Time Analytics in the navigation pane. We can configure the personally identifiable information (PII) identification and redaction settings by choosing Configure advanced settings. For example, we can choose to identify personal data such as email addresses or redact financial data like bank account numbers.

For no extra cost, I can turn on Post-Call Analytics (PCA), so at the end of a call, I get a transcript of the job output into an Amazon Simple Storage Service (Amazon S3) bucket. In doing so, I can use the output from Post-Call Analytics, which is generated by the audio feed, in any processes I already have to produce analytic output generated by the call recordings, such as updating dashboards or producing periodic reports. LCA can now push this post-call analytics data into the latest release of our partner solution (v0.4.0) to deliver analytics visualizations of your live sessions without having to transcode audio a second time.

Unlike LCA (Live call Analytics), which is real-time transcription and analysis of streamed audio, PCA is transcribed and analysis of your recorded calls once your calls are finished. The call is streaming into an LCA app, where it is recorded, transcribed, and analyzed. As you speak into the phone you made the call with, your voice and that of the agents are recorded in real-time and displayed in an automatically scrolling Call Transcripts pane.

Applications

Amazon Transcribe can be used in many common applications, including transcribing customer service calls and creating captions for audio and video content. For the machine, live transcription on stream can be used to transcriber the user’s audio responses on the IVR, voice Bot, or virtual assistant to carry on a conversation between human and machine. Real-time streaming can also be used to look up relevant information from the conversation, then give the contact center agent tips or suggestions on the screen to help resolve the customer’s problem or suggest an upgrade.

This can include stored transcriptions of call center interactions, closed captioned videos, client websites, and other data sources. Real-time analytics: stream audio for transcription and analytics so that any issues, such as an agent failing to repeat an eligibility claim, are instantly addressed. Meeting recaps and analytics – Real-time transcripts and analytics of a meeting enable rapid action after a meeting, for example, identified action items, a summary of the meeting shared, and identified opportunities for sales trainers.

Sales enabling

Transcribing the transcripts into an AI that evaluates the salespersons’ sales pitch and provides recommendations for better closing tactics right after a call or meeting. When authorized users are authenticated, Amazon Cognito sets up a secure graphQL connection to Amazon AppSync APIs and subscribes to receive real-time events, such as new calls and changes in the call status on the Calls List page, as well as new or updated transcript segments, agent assistance messages, and calculated analytics on the Call Details page. The event fires a Lambda function that uses Amazon Simple Queue Service (Amazon SQS) queue to start a new call processing job on AWS Fargate, the container-aware serverless computing service.

Availability and Pricing

Amazon Transcribe Call Analytics with real-time capabilities is available today in the US (N. Virginia, Oregon), Canada (Central), Europe (Frankfurt, London), and Asia Pacific (Seoul, Sydney, Tokyo) and supports US English, British English, Australian English, US Spanish, Canadian French, French, German, Italian, and Brazilian Portuguese.

With Amazon Transcribe Call Analytics, you pay as you go and are billed monthly based on tiered pricing.

For AWS free tier account, you can analyze up to 60 minutes of call audio monthly for free for the first 12 months for Amazon Transcribe Call Analytics, including the new real-time call analytics API.

Conclusion

In conclusion, the addition of real-time analytics during live calls to Amazon Transcribe is a significant development that enhances the capabilities of the speech recognition service. With this feature, users can receive instant insights and feedback during live conversations, allowing them to make adjustments and improve the quality of their interactions. Real-time analytics during live calls is an example of how artificial intelligence and machine learning can enhance human communication and improve customer service, interviews, and other activities involving live conversations. Overall, this feature is a valuable addition to Amazon Transcribe, and it has the potential to transform the way people conduct live conversations and interact with one another in various contexts.

About CloudThat

CloudThat is also the official AWS (Amazon Web Services) Advanced Consulting Partner and Training partner and Microsoft gold partner, helping people develop knowledge of the cloud and help their businesses aim for higher goals using best in industry cloud computing practices and expertise. We are on a mission to build a robust cloud computing ecosystem by disseminating knowledge on technological intricacies within the cloud space. Our blogs, webinars, case studies, and white papers enable all the stakeholders in the cloud computing sphere.

Drop a query if you have any questions regarding Amazon Transcribe and I will get back to you quickly.

To get started, go through our Consultancy page and Managed Services Package that is CloudThat’s offerings.