Machine-learning model for circuit design requirements verification

A non-transitory computer readable medium comprising stored instructions, which when executed by a processor, cause the processor to:

• receive a specification document associated with a circuit design;
• process the specification document to identify a text component, a table component, and a finite state machine (FSM) component;
• parse, by a first parser, the text component to identify a first set of design requirements, parsing the text component including using one or more machine-learning (ML) models to predict whether an identified sentence is likely to be a design requirement, at least one of the one or more ML models trained over a training dataset having (1) positive training samples containing a first set of phrases each of which is labeled as a design requirement, and (2) negative training samples containing a second set of phrases each of which is labeled as not a design requirement;
• parse, by a second parser, the table component to identify a second set of design requirements;
• parse, by a third parser, the FSM component to identify a third set of design requirements;
• and
• provide for display the identified at least one of the first, second, and third sets of design requirements.

A computer system, comprising:

• a processor;
• and
• non-transitory computer readable medium comprising stored instructions, which when executed by the processor, cause the processor to:
• receive a specification document associated with a circuit design;
• process the specification document to identify a text component, a table component, and a finite state machine (FSM) component;
• parse, by a first parser, the text component to identify a first set of design requirements, parsing the text component including using one or more machine-learning (ML) models to predict whether an identified sentence is likely to be a design requirement, at least one of the one or more ML models trained over a training dataset having (1) positive training samples containing a first set of phrases each of which is labeled as a design requirement, and (2) negative training samples containing a second set of phrases each of which is labeled as not a design requirement;
• parse, by a second parser, the table component to identify a second set of design requirements;
• parse, by a third parser, the FSM component to identify a third set of design requirements;
• and
• provide for display the identified at least one of the first, second, and third sets of design requirements.