• Amazon Web Services, Inc. has announced call analytics for the Amazon Chime software development kit, making real-time audio call insights easier and cheaper to gather.
  • Amazon Chime SDK is a collection of components for adding a message, audio, video, and screen-sharing features to online and mobile apps.

Amazon Web Services Inc. recently said it would enable call analytics for the Amazon Chime software development kit, making collecting insights from real-time audio calls more accessible and less expensive.

According to the firm, Amazon Chime SDK call analytics helps developers construct real-time call transcriptions, analyze speech tone, and search for specific speakers. AWS is also enhancing the Amazon Chime SDK section of the AWS Management Console to facilitate incorporating these new features into audio apps.

Amazon Chime SDK is a collection of components for adding message, audio, video, and screen-sharing features to online and mobile apps in real-time. For instance, developers may utilize it to integrate video into a health app, allowing consumers to initiate a video conversation with their doctor straight through the app.

Sébastien Stormacq, Amazon’s principal developer advocate, stated that there are instances in which customers may wish to record and analyze calls conducted through their audio apps. The new voice analytics capability is intended to produce real-time insights based on these calls by extracting emotion from speech cues to detect displeasure and impatience. It operates by analyzing lexical and linguistic information, i.e., what was said, in conjunction with acoustic and tone information, i.e., how it was said.

According to Stormacq, this speech tone analytics will be provided to the customer’s chosen data lake, allowing them to develop dashboards capable of visualizing this data.

Amazon illustrated how this might operate in the financial business, where trading room supervisors frequently record all talks for regulatory purposes. Stormacq stated that voice tone analysis would assist supervisors in monitoring for risk and compliance. In addition, the insights can help traders in increasing their efficiency.

Healthcare, the public sector, telecommunications, and insurance are some businesses that might benefit from voice analysis.

The new speaker search function is intended to discover and identify call participants. Based on a database of known voices, it can identify particular speakers after hearing only a few seconds of their speech. This, according to Stormacq, can assist with duties such as accelerating caller lookup and enhancing call transcripts with identity identification.

Amazon has upgraded the AWS Management Console to make it simpler for developers to utilize these new features. To evaluate a call, developers may simply select the AWS artificial intelligence service, such as voice analytics, Amazon Transcribe, or Amazon Transcribe Call Analytics.

To view these insights, customers may choose Amazon QuickSight or Tableau, whose dashboards can then be incorporated into various apps, wikis, and portals, according to Amazon. Developers may obtain pre-built dashboards as AWS CloudFormation templates rather than developing their own.

Finally, call analytics may be used to produce real-time alerts by publishing specified events to Amazon EventBridge, which then forwards these notifications to the user’s AWS account or any supported third-party application.

Constellation Research Inc.’s Liz Miller told a leading media house that developers would embrace the new call analytics features in Chime since speech and emotion data of this type is essential for training new AI and machine learning models.

Liz Miller said, “This latest SDK gives Chime and AWS customers exactly what they need to turn voice into action. While AI and ML have been announced by almost every major contact center provider, the interesting part of the AWS announcement is that this is really about providing teams building tomorrow’s experiences with the developer toolkits they need to accelerate time to value.”

Amazon stated that Amazon Chime SDK call analytics fees will be usage-based, with customers being charged per minute of audio data examined. Amazon Chime SDK call analytics are presently available in Asia Pacific (Singapore), the US East (Ohio, N. Virginia), and Europe (Frankfurt) regions of AWS, with additional areas to be added in the future.