US 11,455,999 B1 · Detection of off-topic spoken responses using machine learning

Abstract

Data is received that encapsulates a spoken response to a prompt text comprising a string of words. Thereafter, the received data is transcribed into a string of words. The string of words is then compared with a prompt so that a similarity grid representation of the comparison can be generated that characterizes a level of similarity between the string of words in the spoken response and the string of words in the prompt text. The grid representation is then scored using at least one machine learning model. The score indicates a likelihood of the spoken response having been off-topic. Data providing the encapsulated score can then be provided. Related apparatus, systems, techniques and articles are also described.

Claims20

Dependent claims

A computer-implemented method comprising:

receiving data encapsulating a spoken response to a prompt text comprising a string of words;
transcribing the received data into a string of words;
comparing the string of words with the prompt;
generating a similarity grid representation of the comparison that characterizes a level of similarity between the string of words in the spoken response and the string of words in the prompt text, the similarity grid representation being a similarity grid having at least one channel in which each pixel indicates a similarity between pairs of words from the string of words in the spoken response and the string of words in the prompt text;
scoring the grid using at least one machine learning model, the score indicating a likelihood of the spoken response having been off-topic;
and
providing data encapsulating the score.

4 more elements

Dependencies are inferred from the claim text.

Detection of off-topic spoken responses using machine learning

Life of this patent

Detection of off-topic spoken responses using machine learning

Life of this patent

Abstract

Claims20