Ana M. Barbancho

Automatic Transcription of Guitar Chords and Fingering From Audio

This paper proposes a method for extracting the fingering configurations automatically from a recorded guitar performance. 330 different fingering configurations are considered, corresponding to different versions of the major, minor, major 7th, and minor 7th chords played on the guitar fretboard. The method is formulated as a hidden Markov model, where the hidden states correspond to the different fingering configurations and the observed acoustic features are obtained from a multiple fundamental frequency estimator that measures the salience of a range of candidate note pitches within individual time frames. Transitions between consecutive fingerings are constrained by a musical model trained on a database of chord sequences, and a heuristic cost function that measures the physical difficulty of moving from one configuration of finger positions to another. The method was evaluated on recordings from the acoustic, electric, and the Spanish guitar and clearly outperformed a non-guitar-specific reference chord transcription method despite the fact that the number of chords considered here is significantly larger.

Inharmonicity-Based Method for the Automatic Generation of Guitar Tablature

In this paper, a system for the extraction of the tablature of guitar musical pieces using only the audio waveform is presented. The analysis of the inharmonicity relations between the fundamentals and the partials of the notes played is the main process that allows to estimate both the notes played and the string/fret combination that was used to produce that sound. A procedure to analyze chords will also be described. This procedure will also make use of the inharmonicity analysis to find the simultaneous string/fret combinations used to play each chord. The proposed method is suitable for any guitar type: classical, acoustic and electric guitars. The system performance has been evaluated on a series of guitar samples from the RWC instruments database and our own recordings.

Transcription and expressiveness detection system for violin music

In this paper, a transcription system for music played by violin is presented. The transcription system not only detects the pitch and duration of the notes but also identifies successfully the employed technique to play each note: détaché with and without accent and with and without vibrato, pizzicato, tremolo, spiccato and flageolett-töne. The transcription system is based on a combined analysis of the time domain and frequency domain properties of the music signal.

SiPTH: singing transcription based on hysteresis defined on the pitch-time curve

In this paper, we present a method for monophonic singing transcription based on hysteresis defined on the pitch-time curve. This method is designed to perform note segmentation even when the pitch evolution during the same note behaves unstably, as in the case of untrained singers. The selected approach estimates the regions in which the chroma is stable, these regions are classified as voiced or unvoiced according to a decision tree classifier using two descriptors based on aperiodicity and power. Then, a note segmentation stage based on pitch intervals of the sung signal is carried out. To this end, a dynamic averaging of the pitch curve is performed after the beginning of a note is detected in order to roughly estimate the pitch. Deviations of the actual pitch curve with respect to this average are measured to determine the next note change according to a hysteresis process defined on the pitch-time curve. Finally, each note is labeled using three single values: rounded pitch (to semitones), duration and volume. Also, a complete evaluation methodology that includes the definition of different relevant types of errors, measures and a method for the computation of the evaluation measures are presented. The proposed system improves significantly the performance of the baseline approach, and attains results similar to previous approaches.

Fundamental frequency alignment vs. note-based melodic similarity for singing voice assessment

This paper presents a generic approach for automatic singing assessment for basic singing levels. The system provides the user with a set of intonation, rhythm and overall ratings obtained by measuring the similarity of the sung melody and a target performance. Two different similarity approaches are discussed: f0 curve alignment through Dynamic Time Warping (DTW), and singing transcription plus note-level similarity. From these two approaches, we extract different intonation and rhythm similarity measures which are combined through quadratic polynomial regression analysis in order to fit the judgement of 4 trained musicians on 27 performances. The results show that the proposed system is suitable for automatic singing voice rating and that DTW based measures are specially simple and effective for intonation and rhythm assessment.

Automatic melody composition based on a probabilistic model of music style and harmonic rules

Abstract The aim of the present work is to perform a step towards the design of specific algorithms and methods for automatic music generation. A novel probabilistic model for the characterization of music learned from music samples is designed. This model makes use of automatically extracted music parameters, namely tempo, time signature, rhythmic patterns and pitch contours, to characterize music. Specifically, learned rhythmic patterns and pitch contours are employed to characterize music styles. Then, a novel autonomous music compositor that generates new melodies using the model developed will be presented. The methods proposed in this paper take into consideration different aspects related to the traditional way in which music is composed by humans such as harmony evolution and structure repetitions and apply them together with the probabilistic reutilization of rhythm patterns and pitch contours learned beforehand to compose music pieces.

Automatic edition of songs for Guitar Hero/Frets on Fire

In this contribution, an automatic system of song edition for Guitar Hero/Frets on Fire is presented. The system performs three fundamental stages: time analysis, frequency analysis and button allocation. The temporal analysis of the musical signal is used to obtain the rhythmic pattern of the song. The frequency analysis is perform on the basis of aMel filter bank because these filters provide an approximation to the human perception of low and high frequency sounds. The allocation of buttons is done using the rhythmic pattern extracted, the Mel filters outputs and, also, taking into account a selectable difficulty level. The distributions of the buttons obtained with the designed system are similar to the ones found in the predetermined songs of the games.

Human–Computer Interaction and Music

In this chapter, the use of advanced human computer interfaces to create innovative interaction paradigms for music applications (music creation, music manipulation, music games, etc.) is explored. The advances in the design and implementation of sensing technologies have provided the means to create new ways to interact with computers in a more natural way than the conventional computer framework with a mouse and a keyboard. More involving ans immersive experiences can be offered to the user than to these technologies. However, there is no silver bullet: each kind of sensing technology excels at some fields and lack others. Each application will demand its very own selection of sensors and the development of an adequate interaction metaphor. In this chapter, some of the most commonly used technologies for motion sensing are presented with as special focus on the possibilities of a 3D camera sensor (i.e. kinect) with regard to the design of human computer interfaces for musin interaction. We will present our findings in the studies we have conducted using these devices to develop augmented instruments. These include a drumkit stimulator or a virtual theremin. Additionally, the use of this type of interface for other music applications will be discussed. A description of the technical issues that need to be addressed to successfully implement these interaction paradigms is also given.

Unsupervised classification of audio signals by self-organizing maps and bayesian labeling

Audio signal classification consists of extracting some descriptive features from a sound and use them as input in a classifier. Then, the classifier will assign a different label to any different sound class. The classification of the features can be performed in a supervised or unsupervised way. However, unsupervised classification usually supposes a challenge against supervised classification as it has to be performed without any a priori knowledge. In this paper, unsupervised classification of audio signals is accomplished by using a Probabilistic Self-Organizing Map (PSOM) with probabilistic labeling. The hybrid unsupervised classifier presented in this work can achieve higher detection rates than the reached by the unsupervised traditional SOM. Moreover, real audio recordings from clarinet music are used to show the performance of our proposal.

Multirate MC-DS-CDMA transmitter for 4G communications

The scheme of a multirate 4G transmitter based on MC-DS-CDMA is proposed. The system is designed to minimize intercarrier interference (ICI) and multiaccess interference (MAI), taking into account the autocorrelation properties of the CDMA codes assigned to each user on each subcarrier. The system employs two layered spreading codes in a similar way to UMTS, so it will be simple to design terminals that support both systems. The proposed scheme can be employed in both synchronous or asynchronous uplink and downlink. The transmitter is very versatile and it provides multiple possibilities to allow multirate users.