Nadine Kroher

Automatic transcription of flamenco singing from polyphonic music recordings

Automatic note-level transcription is considered one of the most challenging tasks in music information retrieval. The specific case of flamenco singing transcription poses a particular challenge due to its complex melodic progressions, intonation inaccuracies, the use of a high degree of ornamentation, and the presence of guitar accompaniment. In this study, we explore the limitations of existing state of the art transcription systems for the case of flamenco singing and propose a specific solution for this genre: We first extract the predominant melody and apply a novel contour filtering process to eliminate segments of the pitch contour which originate from the guitar accompaniment. We formulate a set of onset detection functions based on volume and pitch characteristics to segment the resulting vocal pitch contour into discrete note events. A quantised pitch label is assigned to each note event by combining global pitch class probabilities with local pitch contour statistics. The proposed system outperforms state of the art singing transcription systems with respect to voicing accuracy, onset detection, and overall performance when evaluated on flamenco singing datasets.

Corpus COFLA: A Research Corpus for the Computational Study of Flamenco Music

Flamenco is a music tradition from Southern Spain that attracts a growing community of enthusiasts around the world. Its unique melodic and rhythmic elements, the typically spontaneous and improvised interpretation, and its diversity regarding styles make this still largely undocumented art form a particularly interesting material for musicological studies. In prior works, it has already been demonstrated that research on computational analysis of flamenco music, despite it being a relatively new field, can provide powerful tools for the discovery and diffusion of this genre. In this article, we present corpusCOFLA, a data framework for the development of such computational tools. The proposed collection of audio recordings and metadata serves as a pool for creating annotated subsets that can be used in development and evaluation of algorithms for specific music information retrieval tasks. First, we describe the design criteria for the corpus creation and then provide various examples of subsets drawn from the corpus. We showcase possible research applications in the context of computational study of flamenco music and give perspectives regarding further development of the corpus.

MIRchiving: Challenges and opportunities of connecting MIR research and digital music archives

This study is a call for action for the music information retrieval (MIR) community to pay more attention to collaboration with digital music archives. The study, which resulted from an interdisciplinary workshop and subsequent discussion, matches the demand for MIR technologies from various archives with what is already supplied by the MIR community. We conclude that the expressed demands can only be served sustainably through closer collaborations. Whereas MIR systems are described in scientific publications, usable implementations are often absent. If there is a runnable system, user documentation is often sparse---posing a huge hurdle for archivists to employ it. This study sheds light on the current limitations and opportunities of MIR research in the context of music archives by means of examples, and highlights available tools. As a basic guideline for collaboration, we propose to interpret MIR research as part of a value chain. We identify the following benefits of collaboration between MIR researchers and music archives: new perspectives for content access in archives, more diverse evaluation data and methods, and a more application-oriented MIR research workflow.

Discovery of repeated vocal patterns in polyphonic audio: A case study on flamenco music

This paper presents a method for the discovery of repeated vocal patterns directly from music recordings. At a first stage, a voice detection algorithm provides a rough segmentation of the recording to vocal parts, based on which an estimate of the average pattern duration is computed. Then, a pattern detector which employs a sequence alignment algorithm is used to yield a ranking of pairs of matches of the detected voiced segments. At a last stage, a clustering algorithm produces the final repeated patterns. Our method was evaluated in the context of flamenco music for which symbolic metadata are very hard to produce, yielding very promising results.

Discovery of Repeated Melodic Phrases in Folk Singing Recordings

In music, repetition is a fundamental concept to establish structure and create temporal relationships. Previous approaches to detecting repetition in music recordings have mainly focused on discovering repeated patterns of variable length and instrumentation at arbitrary locations. In this paper, we present a novel method for the discovery of repeated sung phrases in folk music recordings and, in particular, in oral music traditions, where written scores are usually unavailable. At a first stage, a segmentation algorithm partitions automatically generated note-level transcriptions of the singing melody into sections that correspond to the structural unit of a phrase. A clustering algorithm is then used to form clusters of phrases, where each cluster contains instances of the same melodic content. The clustering algorithm operates on the basis of a distance measure between melodic sequences and, to this end, various melodic distance measures are investigated. A detailed evaluation procedure is used to assess the performance of the algorithm on three different European music traditions and the influence of transcription and segmentation errors is investigated. The proposed system is shown to outperform the state-of-the-art in audio-based approaches to repeated phrase discovery for this task.

Computational Ethnomusicology: A Study of Flamenco and Arab-Andalusian Vocal Music

In this chapter we approach flamenco and Arab-Andalusian vocal music through the analysis of two representative pieces. We apply a hybrid methodology consisting of audio-signal processing to describe and contrast their melodic characteristics followed by musicological analysis. The use of such computational analysis tools complements a musicological-historical study with the aim of supporting the discovery and understanding of the specific characteristics of these musical traditions, their similarities and differences, while offering solutions to more general music information retrieval (MIR) research challenges.

Computing melodic templates in oral music traditions

Abstract The term melodic template or skeleton refers to a basic melody which is subject to variation during a music performance. In many oral music traditions, these templates are implicitly passed throughout generations without ever being formalized in a score. In this work, we introduce a new geometric optimization problem, the spanning tube problem, to approximate a melodic template for a set of labeled performance transcriptions corresponding to a specific style in oral music traditions. Given a set of n piecewise linear functions, we solve the problem of finding a continuous function, f*, and a minimum value, e*, such that, the vertical segment of length 2e* centered at (x, f*(x)) intersects at least p functions (p ≤ n). The method explored here also provide a novel tool for quantitatively assess the amount of melodic variation which occurs across performances.

Audio-Based Melody Categorization: Exploring Signal Representations and Evaluation Strategies

Melody categorization refers to the task of grouping a set of melodies into categories of similar items that originate from the same melodic contour. From a computational perspective, automatic melody categorization is of crucial importance for the automatic organization of databases, as well as for large-scale musicological studies—in particular, in the context of folk music and non-Western music traditions. We investigate methods starting from the raw audio file. For each recording contained in a collection, we extract a pitch sequence representing the main melodic line. We then estimate pairwise similarities and evaluate the discriminative power of the resulting similarity matrix with respect to ground-truth annotations. We propose novel evaluation methodologies, compare melody representations, and explore the potential of our approach in the context of two applications: interstyle and intrastyle categorization of flamenco music and tune-family recognition of folk-song recordings.

Unsupervised singing voice detection using dictionary learning

This paper presents an unsupervised approach to vocal detection in music recordings based on dictionary learning. At a first stage, the recording to be segmented is treated as training data and the K-SVD algorithm is used to learn a dictionary which sparsely represents a short-term feature sequence that has been extracted from the recording. Subsequently, the vectors of the feature sequence are reconstructed based on the learned dictionary and the probability of appearance of the dictionary atoms is estimated. The obtained probability serves to compute the value of a weight function for each frame of the recording. The histogram of this function is then used to estimate a binarization threshold that segments the recording into vocal and non-vocal segments. The performance of the proposed unsupervised method, when evaluated on two datasets of accompanied singing, presents comparable performance to supervised techniques.

Detection of Melodic Patterns in Automatic Transcriptions of Flamenco Singing

The spontaneous expressive interpretation of melodic templates is a fundamental concept in flamenco music. Consequently, the automatic detection of such patterns in music collections sets the basis for a number of challenging analysis and retrieval tasks. We present a novel algorithm for the automatic detection of manually defined melodies within a corpus of automatic transcriptions of flamenco recordings. We evaluate the performance on the example of five characteristic patterns from the fandango de Valverde style and demonstrate that the algorithm is capable of retrieving ornamented instances of query patterns. Furthermore, we discuss limitations, possible extensions and applications of the proposed system.