Amazon Transcribe Call Analytics

Amazon Transcribe Call Analytics is a generative AI-powered API for generating highly accurate call transcripts and extracting conversation insights to improve customer experience and enhance agent and supervisor productivity. The API combines powerful speech-to-text models, large language models (LLMs), and task-specific natural language processing (NLP) models that are trained to understand customer service and sales calls. 

With Amazon Transcribe Call Analytics, you get valuable intelligence such as customer and agent sentiment, call drivers, non-talk time, interruptions, and talk speed. The call categorization capability allows you to classify conversations based on custom criteria such as sentiment, competitive mentions and specific phrases like “not happy,” “poor quality,” and “cancel my subscription”. In addition, you can use generative-AI powered call summarization capability to deliver a concise summary of a customer interaction and capture key components such as why the customer called, how the issue was addressed, and what follow-up actions were identified. Further, the API can help you detect and redact sensitive information such as names, addresses, and credit card information from both the audio and text in real time or post-call. These capabilities help you to improve experiences for customers, agents, and supervisors in your contact centers.

Common use cases for Amazon Transcribe Call Analytics API include real-time agent assist and post-call analytics.

Automatically generate call summaries to help agents focus on providing excellent customer experiences and increase productivity by reducing after-call manual summarization. This capability summarizes an interaction with a caller, capturing key components such as the reason for the call, steps taken to resolve issue, and next steps. Contact center supervisors can review call summaries to quickly understand the context of an interaction without reading the whole transcript while investigating caller issues.

Using the power of ML, you can quickly apply speech-to-text and NLP capabilities during live calls and uncover valuable conversation insights. You can then integrate insights such as customer and agent sentiment, detected issues, and speech characteristics like non-talk time, interruptions, and talk speed into your inbound and outbound call analytics applications. This can help your supervisors more readily identify potential customer issues, agent coaching opportunities, product feedback, and call trends.

Monitor your calls at scale to track compliance with company policies or regulatory requirements. Build and train your own custom categories based on your specified criteria (such as words/phrases or conversation characteristics). For example, you can set up category labels to see what percentage of calls are upsells or account cancellation.

Conversations often contain sensitive customer data such as names, addresses, credit card numbers, and social security numbers. Transcribe Call Analytics helps you identify and redact this information from both the audio and the text.