We are announcing generative artificial intelligence (AI)-powered call summarization in Amazon Transcribe Call Analytics in preview. Powered by Amazon Bedrock, this feature helps businesses improve customer experience, and agent and supervisor productivity by automatically summarizing customer service calls. Amazon Transcribe Call Analytics provides machine learning (ML)-powered analytics that allows contact centers to understand the sentiment, trends, and policy compliance of customer conversations to improve their experience and identify crucial feedback. A single API call is all it takes to extract transcripts, rich insights, and summaries from your customer conversations.
We understand that as a business, you want to maintain an accurate historical record of key conversation points, including action items associated with each conversation. To do this, agents summarize notes after the conversation has ended and enter these in their CRM system, a process that is time-consuming and subject to human error. Now imagine the customer trust erosion that follows when the agent fails to correctly capture and act upon important action items discussed during conversations.
How it works
Starting today, to assist agents and supervisors with the summarization of customer conversations, Amazon Transcribe Call Analytics will generate a concise summary of a contact center interaction that captures key components such as why the customer called, how the issue was addressed, and what follow-up actions were identified. After completing a customer interaction, agents can directly proceed to help the next customer since they don’t have to summarize a conversation, resulting in reduced customer wait times and improved agent productivity. Further, supervisors can review the summary when investigating a customer issue to get a gist of the conversation, without having to listen to the entire call recording or read the transcript.
Exploring Amazon Transcribe Call Analytics in the console
To see how this works visually, I first create an Amazon Simple Storage Service (Amazon S3) bucket in the relevant AWS Region. I then upload the audio file to the S3 bucket.
To create an analytics job that transcribes the audio and provides additional analytics about the conversation that the customer and the agent were having, I go to the Amazon Transcribe Call Analytics console. I select Post-call Analytics in the left hand navigation bar and then choose Create job.
Next I enter a job name making sure to keep the language settings based on the language in the audio file.
In the Amazon S3 URI path, I provide the link to the audio file uploaded in the first screenshot shown in this post.
In Role name, I select Create an IAM role which will have access to the Amazon S3 bucket, then choose Next.
I enable Generative call summarization, and then choose Create job.
After a few minutes, the job’s status will change from In progress to Complete, indicating that it was completed successfully.
Select the job, and the next screen will show the transcript and a new tab, Generative call summarization – preview.
You can also download the transcript to view the analytics and summary.
Now available
Generative call summarization in Amazon Transcribe Call Analytics is available today in English in US East (N. Virginia) and US West (Oregon).
With generative call summarization in Amazon Transcribe Call Analytics, you pay as you go and are billed monthly based on tiered pricing. For more information, see Amazon Transcribe pricing.
Learn more:
– Veliswa
from AWS News Blog https://aws.amazon.com/blogs/aws/amazon-transcribe-call-analytics-adds-new-generative-ai-powered-call-summaries-preview/