It was 2017 when Aws launched Amazon Transcribe, an automatic voice recognition service (ASR, automatic speech recognition) that makes it easy for businesses and developers to add speech-to-text features to any application.
In August, the company announced the availability of Amazon Transcribe Call Analytics, a new feature that allows you to easily extract valuable insights from customer conversations with a single API call.
Any discussion with potential or existing customers
For example, it is important for customer service teams to understand the main reasons why customers call them, and measure satisfaction during these calls.
Similarly, sales departments seek to measure the interest of customers and their reaction to a particular proposal.
Thus, Aws points out, many customers and partners would like to add call analysis capabilities in different applications, regardless of their contact center provider.
Businesses often need to analyse more than phone calls, for example web-based audio and video calls.
So far, customers have typically found a solution to this need by putting together artificial intelligence services and dedicated machine learning models.
But at the same time, they asked Aws to have a simpler solution.
And it is to respond to this customer feedback that Aws has designed Amazon Transcribe Call Analytics, a new addition to Transcribe and ……………..
Based on the Transcribed ASR, Amazon Transcribe Call Analytics adds natural language processing capabilities (NLP) specifically trained on customer calls and optimized to provide highly accurate transcriptions and usable insights
With a simple API call, developers can now easily add call analysis to any application and extract customer insights from conversations without having to build artificial intelligence pipeline and train custom machine learning models.
In addition, it is possible to further process text transcription with other artificial intelligence-based services, such as Amazon Translate, or with custom NLP models built with Amazon SageMaker.