Introduction: Emotion detection has proven to be valuable in biofeedback for the development of assistive technologies, the enhancement of gaming experiences, and advancements in the treatment of mental health issues, among other applications. The objective of this study was to integrate electroencephalogram (EEG), electrocardiogram (ECG), and galvanic skin response (GSR) sensors using the lazypredict library to return classifier models with the highest accuracy in detecting emotions. Method: OpenVibe software; Brain Products' V-Amp amplifier (sample rate 512Hz); a cap with 16 channels placed in temporal, parietal, frontal, and prefrontal regions for EEG; and a BIP2AUX adapter connected to the AUX ports for acquiring ECG signals with three electrodes on the wrists and leg were utilized. The GSR module adapter, with two electrodes on the index and middle fingers, was used for GSR. Each articipant was exposed to 400 emotional stimuli (100 for each emotion—fear, happiness, anger, and sadness) through PsychoPy software. Data were processed using the Python programming language, involving filtering, epoching, epoch selection, feature extraction using discrete wavelet transform (DWT), and normalization. Subsequently, the data were cleaned and classified using the lazypredict library. Results: The classifier models that exhibited the highest accuracy were the Calibrated Classifier CV, the AdaBoost Classifier, and the Decision Tree Classifier. Conclusion: Our findings contribute to advancements in the field of emotion detection, emphasizing the crucial role played by artificial intelligence in the process.