Skip to main content
ELSA
 

Elsa Speech Assessment API

 

Overview​

The ELSA API empowers developers to build AI-driven applications that assess the accuracy and effectiveness of spoken English communication. Using advanced machine learning technology, the API provides a simple and flexible interface for integrating speech evaluation into your application.

The API supports two types of speech analysis:

  • Scripted speech (with transcript)
  • Unscripted speech (without transcript)

For both types, ELSA evaluates key aspects of spoken English, including:

  • Pronunciation
  • Prosody (intonation and rhythm)
  • Fluency

For unscripted recordings, the API additionally evaluates:

  • Grammar
  • Vocabulary

Detailed feedback is provided across these dimensions to help users identify mistakes and improve their speaking skills.

Getting Started​

To begin using the ELSA API, you’ll need an API key to authenticate your requests. If you haven't received one yet, please complete this request form or contact us at [email protected]. An ELSA representative will reach out to assist you.

Scripted API Updates​

The latest version of the Scripted API offers several improvements and extended features. Refer to the Scripted API section of this documentation for a full overview.