METHOD FOR RECOGNIZING EMOTIONS IN SPEECH

Abstract

ABSTRACT A method (100) for recognizing emotions in speech. Further, the method comprising obtaining and pre-processing audio recordings from the Toronto Emotional Speech Set (TESS) dataset to enhance audio quality. Further, the method (100) comprising the steps of extracting acoustic features from the pre-processed audio recordings, including Mel-frequency cepstral coefficients. Further, the method (100) comprising the steps of configuring and training a Long Short-Term Memory (LSTM) neural network using the extracted features and corresponding emotion labels. Further, the method (100) comprising the steps of evaluating the trained LSTM model on a validation dataset using performance metrics including accuracy and F1-score. Further, the method (100) comprising the steps of classifying emotions in new audio samples using the trained LSTM model and analysing the results to assess recognition effectiveness. <>

Specification

Description:
FIELD OF THE DISCLOSURE
[0001] This invention generally relates to the field of advanced materials and, in particular, to a method for developing high-performance biodegradable polymers that are derived from renewable biomass and designed to minimize plastic waste and environmental pollution.
BACKGROUND
[0002] The subject matter discussed in the background section should not be assumed to be prior art merely as a result of its mention in the background section. Similarly, a problem mentioned in the background section or associated with the subject matter of the background section should not be assumed to have been previously recognized in the prior art. The subject matter in the background section merely represents different approaches, which in and of themselves may also correspond to implementations of the claimed technology.
[0003] The increasing concern over environmental pollution and the detrimental effects of conventional plastics has prompted a search for sustainable alternatives. Traditional petr , Claims:1. A method (100) for recognizing emotions in speech, the method comprising the steps of:
obtaining and pre-processing audio recordings from the Toronto Emotional Speech Set (TESS) dataset to enhance audio quality;
extracting acoustic features from the pre-processed audio recordings, including Mel-frequency cepstral coefficients (MFCCs);
configuring and training a Long Short-Term Memory (LSTM) neural network using the extracted features and corresponding emotion labels;
evaluating the trained LSTM model on a validation dataset using performance metrics including accuracy and F1-score; and
classifying emotions in new audio samples using the trained LSTM model and analysing the results to assess recognition effectiveness.

2. The method (100) as claimed in claim 1, wherein the acoustic features extracted in step (b) further include pitch and energy features in addition to the Mel-frequency cepstral coefficients (MFCCs).

Documents