No abstract

We report evidence that long-term memory retains absolute (accurate) features of perceptual events. Specifically, we show that memory for music seems to preserve the absolute tempo of the musical performance. In Experiment 1, 46 subjects sang two different popular songs from memory, and their tempos were compared with recorded versions of the songs. Seventy-two percent of the productions on two consecutive trials came within 8%of the actual tempo, demonstrating accuracy near the perceptual threshold (JND) for tempo. In Experiment 2, a control experiment, we found that folk songs lacking a tempo standard generally have a large variability in tempo; this counters arguments that memory for the tempo of remembered songs is driven by articulatory constraints. The relevance of the present findings to theories of perceptual memory and memory for music is discussed.A fundamental problem facing memory theorists is how to account for two seemingly disparate properties of memory. On the one hand, a rich body of literature suggests that the role of memory is to preserve the gist of experiences; memory functions to formulate general rules and create abstract concepts on the basis of specific exemplars (Posner & Keele, 1970;Rosch, 1975). On the other hand, there is the extensive literature suggesting that memory accurately preserves absolute features of experiences (Brooks, 1978;Jacoby, 1983;Medin & Schaffer, 1978). (For a further discussion of these two perspectives, see McClelland & Rumelhart, 1986.) These perspectives on the function of human memory parallel an old debate in the animal-learning literature about whether animals' internal representations are relational or absolute (Hanson, 1959;Reese, 1968). As with

Model evaluation plays a special role in interactive machine learning (IML) systems in which users rely on their assessment of a model's performance in order to determine how to improve it. A better understanding of what model criteria are important to users can therefore inform the design of user interfaces for model evaluation as well as the choice and design of learning algorithms. We present work studying the evaluation practices of end users interactively building supervised learning systems for real-world gesture analysis problems. We examine users' model evaluation criteria, which span conventionally relevant criteria such as accuracy and cost, as well as novel criteria such as unexpectedness. We observed that users employed evaluation techniquesincluding cross-validation and direct, real-time evaluationnot only to make relevant judgments of algorithms' performance and interactively improve the trained models, but also to learn to provide more effective training data. Furthermore, we observed that evaluation taught users about what types of models were easy or possible to build, and users sometimes used this information to modify the learning problem definition or their plans for using the trained models in practice. We discuss the implications of these findings with regard to the role of generalization accuracy in IML, the design of new algorithms and interfaces, and the scope of potential benefits of incorporating human interaction in the design of supervised learning systems.

Existing audio tools handle the increasing amount of computer audio data inadequately. The typical tape-recorder paradigm for audio interfaces is inflexible and time consuming, especially for large data sets. On the other hand, completely automatic audio analysis and annotation is impossible using current techniques. Alternative solutions are semi-automatic user interfaces that let users interact with sound in flexible ways based on content. This approach offers significant advantages over manual browsing, annotation and retrieval. Furthermore, it can be implemented using existing techniques for audio content analysis in restricted domains. This paper describes MARSYAS, a framework for experimenting, evaluating and integrating such techniques. As a test for the architecture, some recently proposed techniques have been implemented and tested. In addition, a new method for temporal segmentation based on audio texture is described. This method is combined with audio analysis techniques and used for hierarchical browsing, classification and annotation of audio files.

No abstract

No abstract

No abstract

In order to represent musical content, pitch and timing information is utilized in the majority of existing work in Symbolic Music Information Retrieval (MIR). Symbolic representations such as MIDI allow the easy calculation of such information and its manipulation. In contrast, most of the existing work in Audio MIR uses timbral and beat information, which can be calculated using automatic computer audition techniques. In this paper, Pitch Histograms are defined and proposed as a way to represent the pitch content of music signals both in symbolic and audio form. This representation is evaluated in the context of automatic musical genre classification. A multiple-pitch detection algorithm for polyphonic signals is used to calculate Pitch Histograms for audio signals. In order to evaluate the extent and significance of errors resulting from the automatic multiple-pitch detection, automatic musical genre classification results from symbolic and audio data are compared. The comparison indicates that Pitch Histograms provide valuable information for musical genre classification. The results obtained for both symbolic and audio cases indicate that although pitch errors degrade classification performance for the audio case, Pitch Histograms can be effectively used for classification in both cases.