Systems and methods for utilizing contextual information of human speech to generate search parameters

A computer-implemented method, comprising:

• generating search keyword data for a search related to a product, by:
• causing a user device associated with a user to output at least one prompt for a description of a product, wherein each of the at least one prompt is associated with a respective category of keywords in a predetermined list of keywords associated with the product;
• receiving audio data from the user device, the audio data including human speech from the user;
• performing a speech-to-text process on the received audio data to determine text corresponding to, and correlated with, the human speech in the received audio data;
• segmenting the text into one or more discrete conceptual units, such that a respective portion of the human speech is correlated with each of the one or more conceptual units;
• and
• identifying one or more keywords in the predetermined list of keywords that correspond to one or more words in the text, and associating each identified keyword with one or more of the conceptual units having the corresponding one or more words;
• determining, based on the identifying, for each category of keywords, whether the category of keywords consists of keywords unassociated with any of the conceptual units and, in response to the determining, repeating the generating of search keyword data using at least one prompt associated with the category of keywords consisting of keywords unassociated with any of the conceptual units;
• performing a first analysis process on each of the one or more conceptual units to determine a respective first score of each of the one or more conceptual units;
• performing a second analysis process on the respective portion of the human speech correlated with each of the one or more conceptual units to determine a respective second score of each of the one or more conceptual units;
• for each keyword in the predetermined list of keywords that is associated with one or more of the one or more conceptual units, determining a keyword score based on the respective first score and the respective second score of the one or more conceptual units;
• determining at least one product recommendation based on the one or more keywords associated with one or more of the one or more conceptual units and the corresponding keyword score;
• and
• causing the user device to output the at least one product recommendation.

A system comprising:

• a memory storing instructions;
• and
• at least one processor executing the instructions to perform a process including:
• causing a user device associated with a user to output a prompt for a description of a product;
• receiving audio data from the user device, the audio data including human speech from the user;
• performing a speech-to-text process on the received audio data to determine text corresponding to, and correlated with, the human speech in the received audio data;
• segmenting the text into one or more discrete conceptual units, such that a respective portion of the human speech is correlated with each of the one or more conceptual units;
• identifying, from a predetermined list of keywords associated with the product, keywords that correspond to one or more words in the one or more conceptual units, and associating the identified keywords with the one or more conceptual units having the corresponding words;
• performing a sentiment analysis process on each of the one or more conceptual units to determine a respective sentiment score of each of the one or more conceptual units;
• performing an emphasis analysis process on the respective portion of the human speech correlated with each of the one or more conceptual units to determine a respective emphasis score of each of the one or more conceptual units;
• for each keyword in the predetermined list of keywords that is associated with one or more of the one or more conceptual units, determining a keyword score based on the respective sentiment score and the respective emphasis score of the one or more conceptual units;
• determining at least one product recommendation based on the one or more keywords associated with one or more of the one or more conceptual units and the corresponding keyword score;
• and
• causing the user device to output the at least one product recommendation.