TROMPA-MER: an open dataset for personalized music emotion recognition

Abstract

We present a platform and a dataset to help research on Music Emotion Recognition (MER). We developed the Music Enthusiasts platform aiming to improve the gathering and analysis of the so-called “ground truth” needed as input to MER systems. Firstly, our platform involves engaging participants using citizen science strategies and generate music emotion annotations – the platform presents didactic information and musical recommendations as incentivization, and collects data regarding demographics, mood, and language from each participant. Participants annotated each music excerpt with single free-text emotion words (in native language), distinct forced-choice emotion categories, preference, and familiarity. Additionally, participants stated the reasons for each annotation – including those distinctive of emotion perception and emotion induction. Secondly, our dataset was created for personalized MER and contains information from 181 participants, 4721 annotations, and 1161 music excerpts. To showcase the use of the dataset, we present a methodology for personalization of MER models based on active learning. The experiments show evidence that using the judgment of the crowd as prior knowledge for active learning allows for more effective personalization of MER systems for this particular dataset. Our dataset is publicly available and we invite researchers to use it for testing MER systems.

Multimodal Music Datasets? Challenges and Future Goals in Music Processing

“…Multimodal datasets used in emotion-centric tasks, such as CAL500 (38), and AMC (39), combine audio features with emotion annotations. Additional datasets, including those from (40)(41)(42)(43)(44)(45), incorporate labels, lyrics, and participant information. Integrating lyrics with audio data provides additional context, enhancing emotion recognition accuracy.…”

The Emotion-to-Music Mapping Atlas (EMMA): A systematically organized online database of emotionally evocative music excerpts

“…Information about the consistency in emotion ratings of music excerpts is not only important for purposes of emotion induction in experimental studies, but it also plays a critical role in work that uses signal processing and machine (deep) learning methods to characterize or predict musical emotions, such as music emotion recognition (MER; e.g., Gómez-Cañón et al, 2021;Gómez-Cañón et al, 2022;Han et al, 2022;Kim et al, 2010;Panda et al, 2023;Yang & Chen, 2011a). In work seeking to apply machine learning to music and emotion, emotion annotations of music by humans are essential for training machine learning algorithms for classification or recognition of music emotions.…”

“…Characterizations of music-evoked emotion have also relied on concepts from basic emotions theory (Ekman, 1992a), notably happiness, sadness, fear, and anger, sometimes eclectically amplified by concepts that are not part of basic emotions but seem musically plausible to the researchers, such as tenderness or awe (Zentner & Eerola, 2010;Juslin et al, 2016). Examples for datasets with annotations inspired by basic emotions theory include Emotify (Aljanaki et al, 2016), MagnaTagATune (Law et al, 2009), Primary Musical Cues (Eerola, 2016;, SoundTracks (Eerola & Vuoskoski, 2011), and TROMPA [Towards Richer Online Music Public-domain Archives]-MER (Gómez-Cañón et al, 2022). The main limitation of basic emotions theory is that it was conceived to account for survival-type emotions such as anger, fear, or disgust, which Scherer and Zentner (2008) characterized as "utilitarian" as opposed to "aesthetic emotions".…”

A Comparative Study on Choir Teaching Based on Diversified Models in Chinese and Foreign Universities under the Background of Information Technology

“…In the teaching process, more methods such as practicing vocals, studying music scores, and engaging in gatherings are used to improve students' basic level. At the same time, attention is also paid to academic research on choral singing and the study of choral history, cultural background, and musical personality [9]. Chinese universities focus on basic practice and learning in their teaching content, which is also a weakness compared to universities in other countries.…”