Cover Song Identification Research Articles

Cochlear pitch class profile for cover song identification

Pitch class profile (PCP), which can represent the harmonic progression of a piece of music very well, is one of the widely used audio features for cover version identification. In this letter, we describe a novel procedure that enhances PCP by substantially boosting the degree of instrumental accompaniment invariance without degrading the feature’s discriminative power. Our idea is based on the assumption that human ear can identify a cover of a pop song based on their singing voice quickly and easily. So, we combine two concepts from psychoacoustics: (i) time-varying loudness contour and (ii) critical band, which have been used in speech recognition field successfully, with the conventional PCP descriptor to enhance its discriminative power. Since the CPCPs aim at a representation of singing voice, they may also obtain improved performance (as compared to conventional PCPs) when applied to a cappella singing recordings. Experimental results demonstrate that the resulting PCP feature, called cochlear pitch class profile (CPCP), outperforms conventional PCP feature in the context of pop cover song identification application.

Identifying Cover Songs Using Information-Theoretic Measures of Similarity

This paper investigates methods for quantifying similarity between audio signals, specifically for the task of of cover song detection. We consider an information-theoretic approach, where we compute pairwise measures of predictability between time series. We compare discrete-valued approaches operating on quantised audio features, to continuous-valued approaches. In the discrete case, we propose a method for computing the normalised compression distance, where we account for correlation between time series. In the continuous case, we propose to compute information-based measures of similarity as statistics of the prediction error between time series. We evaluate our methods on two cover song identification tasks using a data set comprised of 300 Jazz standards and using the Million Song Dataset. For both datasets, we observe that continuous-valued approaches outperform discrete-valued approaches. We consider approaches to estimating the normalised compression distance (NCD) based on string compression and prediction, where we observe that our proposed normalised compression distance with alignment (NCDA) improves average performance over NCD, for sequential compression algorithms. Finally, we demonstrate that continuous-valued distances may be combined to improve performance with respect to baseline approaches. Using a large-scale filter-and-refine approach, we demonstrate state-of-the-art performance for cover song identification using the Million Song Dataset.

Characterization and exploitation of community structure in cover song networks

The use of community detection algorithms is explored within the framework of cover song identification, i.e. the automatic detection of different audio renditions of the same underlying musical piece. Until now, this task has been posed as a typical query-by-example task, where one submits a query song and the system retrieves a list of possible matches ranked by their similarity to the query. In this work, we propose a new approach which uses song communities (clusters, groups) to provide more relevant answers to a given query. Starting from the output of a state-of-the-art system, songs are embedded in a complex weighted network whose links represent similarity (related musical content). Communities inside the network are then recognized as groups of covers and this information is used to enhance the results of the system. In particular, we show that this approach increases both the coherence and the accuracy of the system. Furthermore, we provide insight into the internal organization of individual cover song communities, showing that there is a tendency for the original song to be central within the community. We postulate that the methods and results presented here could be relevant to other query-by-example tasks.

Predictability of Music Descriptor Time Series and its Application to Cover Song Detection

Intuitively, music has both predictable and unpredictable components. In this paper, we assess this qualitative statement in a quantitative way using common time series models fitted to state-of-the-art music descriptors. These descriptors cover different musical facets and are extracted from a large collection of real audio recordings comprising a variety of musical genres. Our findings show that music descriptor time series exhibit a certain predictability not only for short time intervals, but also for mid-term and relatively long intervals. This fact is observed independently of the descriptor, musical facet and time series model we consider. Moreover, we show that our findings are not only of theoretical relevance but can also have practical impact. To this end we demonstrate that music predictability at relatively long time intervals can be exploited in a real-world application, namely the automatic identification of cover songs (i.e., different renditions or versions of the same musical piece). Importantly, this prediction strategy yields a parameter-free approach for cover song identification that is substantially faster, allows for reduced computational storage and still maintains highly competitive accuracies when compared to state-of-the-art systems.

Cross recurrence quantification for cover song identification

There is growing evidence that nonlinear time series analysis techniques can be used to successfully characterize, classify, or process signals derived from real-world dynamics even though these are not necessarily deterministic and stationary. In the present study, we proceed in this direction by addressing an important problem our modern society is facing, the automatic classification of digital information. In particular, we address the automatic identification of cover songs, i.e. alternative renditions of a previously recorded musical piece. For this purpose, we here propose a recurrence quantification analysis measure that allows the tracking of potentially curved and disrupted traces in cross recurrence plots (CRPs). We apply this measure to CRPs constructed from the state space representation of musical descriptor time series extracted from the raw audio signal. We show that our method identifies cover songs with a higher accuracy as compared to previously published techniques. Beyond the particular application proposed here, we discuss how our approach can be useful for the characterization of a variety of signals from different scientific disciplines. We study coupled Rössler dynamics with stochastically modulated mean frequencies as one concrete example to illustrate this point.

Chroma Binary Similarity and Local Alignment Applied to Cover Song Identification

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.

Melody-based retrieval in audio collections

Mid-level representations are increasingly used to bridge the gap between high-level (semantic) and low-level audio representations. A mid-level representation that integrates melodic and rhythmic aspects of a music signal is introduced. The representation is formed by first performing multi-pitch detection on consecutive audio frames, and then searching for dominant melodic lines within the detected pitches. Beat-tracking is also performed to yield a beat-synchronous representation, independent of tempo variations. The representation is used as the basis for a melody-based audio retrieval system. An approximate nearest neighbor search algorithm is employed to compare sections of the mid-level representation of the query to indexed sections of mid-level representations of songs in a collection. Symbolic queries may also be used. Results are ranked according to the number of matched sections. A locality sensitive hashing function for cosine similarity is used for indexing, and the point location in equal balls algorithm for searching. Retrieval was tested on the cover song identification task, where the goal is to retrieve different interpretations of a song in a collection. Results on a collection of 2,400 songs are presented. [Work supported by ARRS.]