Browse Prior Art Database

System and method to add emoticons against words containing sentiment or emotion that can be integrated to a "speech to text converter (STT)", based on accents / pronunciation used by the speaker.

IP.com Disclosure Number: IPCOM000249511D
Publication Date: 2017-Mar-02
Document File: 3 page(s) / 59K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a System that converts the communicated audio message, via social media, into text with proper emoticons specific to word/ part of speech based on the emotional state of communicator, usage of accented pronunciation and a "Repository Model for Comprehensive emotional sentiment and Pronunciation map". System and method to add emotional tone/accent to "text to speech converter (TTS)", based on comprehensive psycholinguistic profile of the user.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 49% of the total text.

1

System and method to add emoticons against words containing sentiment or emotion that can be integrated to a "speech to text converter (STT)", based on accents / pronunciation used by the speaker .

Problem Solved :

During social communication converting text to speech and speech to text has become a practical need now a days. There are tools and methods already available for converting speech communication onto text and vice versa that can be integrated seamlessly with social Media and other cognitive systems. However most of those converters speech to text in a flat way without considering the emotional state of the communicator. So if a recipient of the communication is willing to read the audio message with emotion, meant for him/her and unable to hear to the audio, it is not possible to understand the emotional state of the communicator correctly just by looking at the monotonous/stereotype text converted by the speech recognition system or speech to text converter.

The system converts the communicated audio message, via social media, into text with proper emoticons specific to word/ part of speech based on the emotional state of communicator, usage of accented pronunciation and a “Repository Model for Comprehensive emotional sentiment and Pronunciation map”. System and method to add emotional tone/accent to "text to speech converter (TTS)", based on comprehensive psycholinguistic profile of the user.

Advantage :

1) Correct conversion of “text synchronized with emoticons”, from speech, provides any recipient the knowledge about the correct sentiment and emotion of the communicator. 2) If the recipient is able to understand the status of emotion of the sender by reading the converted communicated text synchronized with emoticons, he/she can take the appropriate decision/judgement/action as may be necessary. Without the knowledge of state of emotion this is not possible.

Figure-1 :

2

Solution Details : Below are the steps to implement the system that is represented in above Figure

1) System that Records the voice and break down into word wave fragments (audio stream) which represent words/phrase for the text.

2) System that compare and finds out the corresponding word/phrase against each wave fragments.

3) System that sends communicated message in the form of verbal, written and expressed body language to “System and method to capture all communications made during project execution via verbal, written and body language and analyzes the same to build a comprehensive Psycholinguistic – Ecosystem for an enterprise and recommends user about better communication approach for any forth coming discussion”. This produces a comprehensive Psycholinguistic profile.

4) System that gets the Accent features(Tone, pitch etc.) from the wave fragments for each corresponding word/phrase.

5) System that invokes the Cognitive service and receives the Directional sentiment of each part of speech of the received text.

6) System that calculates Comprehensive e...