Emotion-Aware Music Dataset with Multimodal Audio Features

Emotion-Aware Music Sentiment Dataset provides multimodal audio features and contextual metadata for emotion-based music AI. The dataset originates from Kaggle, though specific details on volume, authorship, and recency are unavailable.

Use Cases

• Train models to predict music sentiment labels from extracted audio features like spectral centroid or MFCCs.
• Develop multimodal classifiers that fuse contextual metadata with audio feature vectors for emotion recognition.
• Analyze correlations between specific audio feature sets and human-perceived emotional states in music tracks.

Strengths

• Dataset is explicitly designed for multimodal emotion analysis, combining audio features with contextual metadata.

Limitations

• Dataset size, row count, and specific feature dimensions are unknown, making suitability for model training difficult to assess.
• Potential limitations like class imbalance, label noise, or geographic bias in music selection cannot be verified without sample data.

Provenance

Source

Kaggle