Speech Emotion Recognition Using Self-Supervised Features