Tollywood Emotions: Annotation of Valence-Arousal in Telugu Song Lyrics
Preprint
Abstract
Emotion recognition from a given music track has heavily relied on acoustic features, social tags, and metadata but is seldom focused on lyrics. There are no datasets of Indian language songs that contain both valence and arousal manual ratings of lyrics. We present a new manually annotated dataset of Telugu songs' lyrics collected from Spotify with valence and arousal annotated on a discrete scale. A fairly high inter-annotator agreement was observed for both valence and arousal. Subsequently, we create two music emotion recognition models by using two classification techniques to identify valence, arousal and respective emotion quadrant from lyrics. Support vector machine (SVM) with term frequency-inverse document frequency (TF-IDF) features and fine-tuning the pre-trained XLMRoBERTa (XLM-R) model were used for valence, arousal and quadrant classification tasks. Fine-tuned XLMRoBERTa performs better than the SVM by improving macro-averaged F1-scores of 54.69%, 67.61%, 34.13% to 77.90%, 80.71% and 58.33% for valence, arousal and quadrant classifications, respectively, on 10-fold cross-validation. In addition, we compare our lyrics annotations with Spotify's annotations of valence and energy (same as arousal), which are based on entire music tracks. The implications of our findings are discussed. Finally, we make the dataset publicly available with lyrics, annotations and Spotify IDs.Keywords
All Related Versions
This publication has 10 references indexed in Scilit:
- Cultural differences in music features across Taiwanese, Japanese and American marketsPeerJ Computer Science, 2021
- A Sentiment Analysis System for the Hindi Language by Integrating Gated Recurrent Unit with Genetic AlgorithmThe International Arab Journal of Information Technology, 2020
- Static and Dynamic Measures of Active Music Listening as Indicators of Depression Risk2020
- Emotions of music listening in Finland and in India: Comparison of an individualistic and a collectivistic culturePsychology of Music, 2020
- Unsupervised Cross-lingual Representation Learning at Scale2020
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding2019
- Emotionally-Relevant Features for Classification and Regression of Music LyricsIEEE Transactions on Affective Computing, 2016
- Understanding inverse document frequency: on theoretical arguments for IDFJournal of Documentation, 2004
- Sentiment Classification of Reviews Using SentiWordNet
- Japanese Opinion Surveys: The Special Need and the Special DifficultiesPublic Opinion Quarterly, 1945