Claims that generative AI enhances human creativity rely on individualistic notions that obscure the socio-technical conditions shaping creative practice. This framing enables what we call creativity-washing—the mobilisation of creativity discourse to legitimise corporate control over creative labour—and narrows evaluation to copyright compliance and user experience. We propose a multi-relational evaluative framework examining generative AI across scales: micro-level experience (postphenomenology), meso-level networks of agency (actor-network theory), and macro-level cultural and economic conditions (materialist-relational perspective). The framework surfaces concerns that single-dimension metrics, such as openness rankings, usability studies, and output benchmarks, render invisible. We illustrate the framework through a comparative analysis of two generative audio systems, Stable Audio Open and RAVE. The analysis yields findings at each scale, including the observation that openness at the point of access can coexist with enclosure at the point of governance.

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.

Our previous research showed promising results when transferring features learned from speech to train emotion recognition models for music. In this context, we implemented a denoising autoencoder as a pretraining approach to extract features from speech in two languages (English and Mandarin). From that, we performed transfer and multi-task learning to predict classes from the arousal-valence space of music emotion. We tested and analyzed intra-linguistic and cross-linguistic settings, depending on the language of speech and lyrics of the music. This paper presents additional investigation on our approach, which reveals that: (1) performing pretraining with speech in a mixture of languages yields similar results than for specific languages - the pretraining phase appears not to exploit particular language features, (2) the music in Mandarin dataset consistently results in poor classification performance - we found low agreement in annotations, and (3) novel methodologies for representation learning (Contrastive Predictive Coding) may exploit features from both languages (i.e., pretraining on a mixture of languages) and improve classification of music emotions in both languages. From this study we conclude that more research is still needed to understand what is actually being transferred in these type of contexts.

Intelligent technologies have become an essential part of music systems and applications. This is evidenced by today's omnipresence of digital online music stores and streaming services, which rely on music recommenders, automatic playlist generators, and music browsing interfaces. A large amount of research leading to intelligent music applications deals with the extraction of musical and acoustic information directly from the audio signal using signal processing techniques. Other strategies exploit contextual aspects of music, not present in the signal, for example, community meta-data and trails of user interaction, as found, for instance, on social media platforms. In this editorial, we discuss the notion of “intelligent music system” and give an overview of the papers selected to this special issue.

We present Essentia 2.0, an open-source C++ library for audio analysis and audio-based music information retrieval released under the Affero GPL license. It contains an extensive collection of reusable algorithms which implement audio input/output functionality, standard digital signal processing blocks, statistical characterization of data, and a large set of spectral, temporal, tonal and high-level music descriptors. The library is also wrapped in Python and includes a number of predefined executable extractors for the available music descriptors, which facilitates its use for fast prototyping and allows setting up research experiments very rapidly. Furthermore, it includes a Vamp plugin to be used with Sonic Visualiser for visualization purposes. The library is cross-platform and currently supports Linux, Mac OS X, and Windows systems. Essentia is designed with a focus on the robustness of the provided music descriptors and is optimized in terms of the computational cost of the algorithms. The provided functionality, specifically the music descriptors included in-the-box and signal processing algorithms, is easily expandable and allows for both research experiments and development of large-scale industrial applications.

In this paper we describe a system for content-based retrieval of audio clips from a large unstructured database. The system allows users to devise their own sound taxonomies for organizing sounds. An active learning algorithm for Support Vector Machines (SVM) is used for reducing the effort of annotating sounds in the database.

Measuring music similarity is essential for multimedia retrieval. For music items, this task can be regarded as obtaining a suitable distance measurement between songs defined on a certain feature space. In this paper, we propose three of such distance measures based on the audio content: first, a low-level measure based on tempo-related description; second, a high-level semantic measure based on the inference of different musical dimensions by support vector machines. These dimensions include genre, culture, moods, instruments, rhythm, and tempo annotations. Third, a hybrid measure which combines the above-mentioned distance measures with two existing low-level measures: a Euclidean distance based on principal component analysis of timbral, temporal, and tonal descriptors, and a timbral distance based on single Gaussian Mel-frequency cepstral coefficient (MFCC) modeling. We evaluate our proposed measures against a number of baseline measures. We do this objectively based on a comprehensive set of music collections, and subjectively based on listeners' ratings. Results show that the proposed methods achieve accuracies comparable to the baseline approaches in the case of the tempo and classifier-based measures. The highest accuracies are obtained by the hybrid distance. Furthermore, the proposed classifier-based approach opens up the possibility to explore distance measures that are based on semantic notions.

Studying the ways to recommend music to a user is a central task within the music information research community. From a content-based point of view, this task can be regarded as obtaining a suitable distance measurement between songs defined on a certain feature space. We propose two such distance measures. First, a low-level measure based on tempo-related aspects, and second, a high-level semantic measure based on regression by support vector machines of different groups of musical dimensions such as genre and culture, moods and instruments, or rhythm and tempo. We evaluate these distance measures against a number of state-of-the-art measures objectively, based on 17 ground truth musical collections, and subjectively, based on 12 listeners’ ratings. Results show that, in spite of being conceptually different, the proposed methods achieve comparable or even higher performance than the considered baseline approaches. Furthermore, they open up the possibility to explore distance metrics that are based on truly semantic notions.

A causal system to represent a stream of music into musical events, and to generate further expected events, is presented. Starting from an auditory front-end that extracts low-level (i.e. MFCC) and mid-level features such as onsets and beats, an unsupervised clustering process builds and maintains a set of symbols aimed at representing musical stream events using both timbre and time descriptions. The time events are represented using inter-onset intervals relative to the beats. These symbols are then processed by an expectation module using Predictive Partial Match, a multiscale technique based on N-grams. To characterise the ability of the system to generate an expectation that matches both ground truth and system transcription, we introduce several measures that take into account the uncertainty associated with the unsupervised encoding of the musical sequence. The system is evaluated using a subset of the ENST-drums database of annotated drum recordings. We compare three approaches to combine timing (when) and timbre (what) expectation. In our experiments, we show that the induced representation is useful for generating expectation patterns in a causal way.

In this paper we present a study on music mood classification using audio and lyrics information. The mood of a song is expressed by means of musical features but a relevant part also seems to be conveyed by the lyrics. We evaluate each factor independently and explore the possibility to combine both, using natural language processing and music information retrieval techniques. We show that standard distance-based methods and latent semantic analysis are able to classify the lyrics significantly better than random, but the performance is still quite inferior to that of audio-based techniques. We then introduce a method based on differences between language models that gives performances closer to audio-based classifiers. Moreover, integrating this in a multimodal system (audio+text) allows an improvement in the overall performance. We demonstrate that lyrics and audio information are complementary, and can be combined to improve a classification system.

This paper presents two methods, named Item- and User-centric, to evaluate the quality of novel recommendations. The former method focuses on analyzing the item-based recommendation network. The aim is to detect whether the network topology has any pathology that hinders novel recommendations. The latter, user-centric evaluation, aims at measuring users' perceived quality of novel, previously unknown, recommendations.The results of the experiments, done in the music recommendation context, show that last.fm social recommender, based on collaborative filtering, is prone to popularity bias. This has direct consequences on the topology of the item-based recommendation network. Pure audio content-based methods (CB) are not affected by popularity. However, a user-centric experiment done with 288 subjects shows that even though a social-based approach recommends less novel items than our CB, users' perceived quality is better than those recommended by a pure CB method.

In this paper, we optimize and evaluate computational models of similarity for sounds from the same instrument class. We investigate four instrument classes: bass drums, snare drums, high-pitched toms, and low-pitched toms. We evaluate two similarity models: one is defined in the ISO/IEC MPEG-7 standard, and the other is based on auditory images. For the second model, we study the impact of various parameters. We use data from listening tests, and instrument class labels to evaluate the models. Our results show that the model based on auditory images yields a very high average correlation with human similarity ratings and clearly outperforms the MPEG-7 recommendation. The average correlations range from 0.89-0.96 depending on the instrument class. Furthermore, our results indicate that instrument class data can be used as alternative to data from listening tests to evaluate sound similarity models.

We present a new technique for audio signal comparison based on tonal subsequence alignment and its application to detect cover versions (i.e., different performances of the same underlying musical piece). Cover song identification is a task whose popularity has increased in the music information retrieval (MIR) community along in the past, as it provides a direct and objective way to evaluate music similarity algorithms. This paper first presents a series of experiments carried out with two state-of-the-art methods for cover song identification. We have studied several components of these (such as chroma resolution and similarity, transposition, beat tracking or dynamic time warping constraints), in order to discover which characteristics would be desirable for a competitive cover song identifier. After analyzing many cross-validated results, the importance of these characteristics is discussed, and the best performing ones are finally applied to the newly proposed method. Multiple evaluations of this one confirm a large increase in identification accuracy when comparing it with alternative state-of-the-art approaches.

We report on a method of automatic extraction of a metrical attribute from percussive music audio signals: the smallest rhythmic pulse, called the ìtickî. The relevance of use of this feature in the framework of subsequent analyses is discussed and evaluated.