Understanding emotion in music is an important task in music information retrieval with a wide range of applications in music recommendation, playlist generation, and music therapy. However, the lack of large, public datasets of music and emotion is a major roadblock for researchers. In this paper, we leverage an existing music resource − concert program notes − to create a novel dataset of emotion labels for classical music for research in music emotion recognition. We collect program notes from the New York Philharmonic Society’s archive as unorganized documents. We use a large language model, LLaMA 3.1, with task-specific prompting to extract program notes for corresponding musical works. From a total of 9,569 documents, we extracted 31,966 program notes and annotated with emotion values determined by the textual content. We use a dataset of emotion values of English words to estimate valence, arousal, and dominance scores for each program note. Finally, we present an emotion recognition model using another LLM, Longformer, trained on our dataset to estimate valence, arousal, and dominance scores of the program notes. We achieved Pearson’s correlation of 0.877, 0.823, and 0.766 for valence, arousal, and dominance, respectively.