Pop Music Highlighter: Marking the Emotion Keypoints
| Autor: | Yi-Hsuan Yang, Szu-Yu Chou, Yu-Siang Huang |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: |
FOS: Computer and information sciences
lcsh:M1-5000 Sound (cs.SD) Computer Science - Artificial Intelligence Computer science Emotion classification convolutional neural network 02 engineering and technology computer.software_genre Convolutional neural network highlight extraction Computer Science - Sound Popular music Audio and Speech Processing (eess.AS) FOS: Electrical engineering electronic engineering information engineering 0202 electrical engineering electronic engineering information engineering Feature (machine learning) Set (psychology) Music thumbnailing lcsh:Music lcsh:T58.5-58.64 Heuristic business.industry lcsh:Information technology structure analysis Multimedia (cs.MM) Artificial Intelligence (cs.AI) Recurrent neural network 020201 artificial intelligence & image processing Repetition (music) Artificial intelligence business chorus detection attention mechanism computer Computer Science - Multimedia Natural language processing Electrical Engineering and Systems Science - Audio and Speech Processing |
| Zdroj: | Transactions of the International Society for Music Information Retrieval, Vol 1, Iss 1, Pp 68-78 (2018) |
| ISSN: | 2514-3298 |
| Popis: | The goal of music highlight extraction is to get a short consecutive segment of a piece of music that provides an effective representation of the whole piece. In a previous work, we introduced an attention-based convolutional recurrent neural network that uses music emotion classification as a surrogate task for music highlight extraction, for Pop songs. The rationale behind that approach is that the highlight of a song is usually the most emotional part. This paper extends our previous work in the following two aspects. First, methodology-wise we experiment with a new architecture that does not need any recurrent layers, making the training process faster. Moreover, we compare a late-fusion variant and an early-fusion variant to study which one better exploits the attention mechanism. Second, we conduct and report an extensive set of experiments comparing the proposed attention-based methods against a heuristic energy-based method, a structural repetition-based method, and a few other simple feature-based methods for this task. Due to the lack of public-domain labeled data for highlight extraction, following our previous work we use the RWC POP 100-song data set to evaluate how the detected highlights overlap with any chorus sections of the songs. The experiments demonstrate the effectiveness of our methods over competing methods. For reproducibility, we open source the code and pre-trained model at https://github.com/remyhuang/pop-music-highlighter/. Comment: Transactions of the ISMIR vol. 1, no. 1 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|