A fine-tuned Wav2Vec2 model for dimensional speech emotion recognition using the MSP-Podcast dataset.