Method, system, and device for performing real-time sentiment modulation in conversation systems

A system for performing real-time sentiment modulation in a conversation system, the system comprising:

• a processor;
• and
• a memory communicatively coupled to the processor, wherein the memory stores processor instructions, which, on execution, cause the processor to:
• generate an impact table comprising a plurality of sentiment vectors for sentiments associated with a plurality of sentences and a plurality of emotion vectors for emotions associated with the plurality of sentences, wherein a sentiment vector from the plurality of sentiment vectors comprises a sentiment weight assigned to each word of an associated sentence from the plurality of sentences, and wherein an emotion vector from the plurality of emotion vectors comprises an emotion weight assigned to each word of the associated sentence from the plurality of sentences;
• generate for each of the plurality of sentences, a dependency vector based on the associated sentiment vector from the plurality of sentiment vectors and the associated emotion vector from the plurality of emotion vectors;
• stack the dependency vector generated for each of the plurality of sentences to generate a waveform representing variance in sentiment and emotions across words within the plurality of sentences;
• alter at least one portion of the waveform based on a desired emotional output to generate a reshaped waveform, wherein variance in sentiment for the at least one portion is above a predefined threshold;
• generate a set of rephrased sentences for a set of sentences associated with the at least one portion, based on the reshaped waveform, the set of sentences, and a user defined sentiment output;
• and
• generate a prosody for the set of rephrased sentences and a remaining plurality of sentences based on the reshaped waveform, wherein generating the prosody comprises:
• creating a composite sentiment profile based on a set of user requirements;
• and
• averaging a sentiment signal corresponding to the reshaped waveform to generate a prosody waveform based on the composite sentiment profile, wherein the prosody waveform is transposed into the prosody.

A non-transitory computer-readable storage medium for automating repetitive task on a user interface, having stored thereon, a set of computer-executable instructions causing a computer comprising one or more processors to perform steps comprising:

• generating an impact table comprising a plurality of sentiment vectors for sentiments associated with a plurality of sentences and a plurality of emotion vectors for emotions associated with the plurality of sentences, wherein a sentiment vector from the plurality of sentiment vectors comprises a sentiment weight assigned to each word of an associated sentence from the plurality of sentences, and wherein an emotion vector from the plurality of emotion vectors comprises an emotion weight assigned to each word of the associated sentence from the plurality of sentences;
• generating for each of the plurality of sentences, a dependency vector based on the associated sentiment vector from the plurality of sentiment vectors and the associated emotion vector from the plurality of emotion vectors;
• stacking the dependency vector generated for each of the plurality of sentences to generate a waveform representing variance in sentiment and emotions across words within the plurality of sentences;
• altering at least one portion of the waveform based on a desired emotional output to generate a reshaped waveform, wherein variance in sentiment for the at least one portion is above a predefined threshold;
• generating a set of rephrased sentences for a set of sentences associated with the at least one portion, based on the reshaped waveform, the set of sentences, and a user defined sentiment output;
• and
• generating a prosody for the set of rephrased sentences and a remaining plurality of sentences based on the reshaped waveform, wherein generating the prosody comprises:
• creating a composite sentiment profile based on a set of user requirements;
• and
• averaging a sentiment signal corresponding to the reshaped waveform to generate a prosody waveform based on the composite sentiment profile, wherein the prosody waveform is transposed into the prosody.