Preeti Rao

Vocal Melody Extraction in the Presence of Pitched Accompaniment in Polyphonic Music

Melody extraction algorithms for single-channel polyphonic music typically rely on the salience of the lead melodic instrument, considered here to be the singing voice. However the simultaneous presence of one or more pitched instruments in the polyphony can cause such a predominant-F0 tracker to switch between tracking the pitch of the voice and that of an instrument of comparable strength, resulting in reduced voice-pitch detection accuracy. We propose a system that, in addition to biasing the salience measure in favor of singing voice characteristics, acknowledges that the voice may not dominate the polyphony at all instants and therefore tracks an additional pitch to better deal with the potential presence of locally dominant pitched accompaniment. A feature based on the temporal instability of voice harmonics is used to finally identify the voice pitch. The proposed system is evaluated on test data that is representative of polyphonic music with strong pitched accompaniment. Results show that the proposed system is indeed able to recover melodic information lost to its single-pitch tracking counterpart, and also outperforms another state-of-the-art melody extraction system designed for polyphonic music.

Speech enhancement in nonstationary noise environments using noise properties

Abstract Traditional short-time spectral attenuation (STSA) speech enhancement algorithms are ineffective in the presence of highly nonstationary noise due to difficulties in the accurate estimation of the local noise spectrum. With a view to improve the speech quality in the presence of random noise bursts, characteristic of many environmental sounds, a simple post-processing scheme is proposed that can be applied to the output of an STSA speech enhancement algorithm. The post-processing algorithm is based on using spectral properties of the noise in order to detect noisy time–frequency regions which are then attenuated using a SNR-based rule. A suitable suppression rule is developed that is applied to the detected noisy regions so as to achieve significant reduction of noise with minimal speech distortion. The post-processing method is evaluated in the context of two well-known STSA speech enhancement algorithms and experimental results demonstrating improved speech quality are presented for a data set of real noise samples.

Rāga Recognition based on Pitch Distribution Methods

Abstract Rāga forms the melodic framework for most of the music of the Indian subcontinent. Thus automatic rāga recognition is a fundamental step in the computational modelling of the Indian art-music traditions. In this work, we investigate the properties of rāga and the natural processes by which people identify it. We bring together and discuss the previous computational approaches to rāga recognition correlating them with human techniques, in both Karṇāṭaka (south Indian) and Hindustānī (north Indian) music traditions. The approaches which are based on first-order pitch distributions are further evaluated on a large comprehensive dataset to understand their merits and limitations. We outline the possible short and mid-term future directions in this line of work.

Classification of Melodic Motifs in Raga Music with Time-series Matching

Abstract Ragas are characterized by their melodic motifs or catch phrases that constitute strong cues to the raga identity for both the performer and the listener, and therefore are of great interest in music retrieval and automatic transcription. While the characteristic phrases, or pakads, appear in written notation as a sequence of notes, musicological rules for interpretation of the phrase in performance in a manner that allows considerable creative expression, while not transgressing raga grammar, are not explicitly defined. In this work, machine learning methods are used on labelled databases of Hindustani and Carnatic vocal audio concerts to obtain phrase classification on manually segmented audio. Dynamic time warping and HMM based classification are applied on time series of detected pitch values used for the melodic representation of a phrase. Retrieval experiments on raga-characteristic phrases show promising results while providing interesting insights on the nature of variation in the surface realization of raga-characteristic motifs within and across concerts.

A measure for predicting audibility discrimination thresholds for spectral envelope distortions in vowel sounds

Both in speech synthesis and in sound coding it is often beneficial to have a measure that predicts whether, and to what extent, two sounds are different. This paper addresses the problem of estimating the perceptual effects of small modifications to the spectral envelope of a harmonic sound. A recently proposed auditory model is investigated that transforms the physical spectrum into a pattern of specific loudness as a function of critical band rate. A distance measure based on the concept of partial loudness is presented, which treats detectability in terms of a partial loudness threshold. This approach is adapted to the problem of estimating discrimination thresholds related to modifications of the spectral envelope of synthetic vowels. Data obtained from subjective listening tests using a representative set of stimuli in a 3IFC adaptive procedure show that the model makes reasonably good predictions of the discrimination threshold. Systematic deviations from the predicted thresholds may be related to individual differences in auditory filter selectivity. The partial loudness measure is compared with previously proposed distance measures such as the Euclidean distance between excitation patterns and between specific loudness applied to the same experimental data. An objective test measure shows that the partial loudness measure and the Euclidean distance of the excitation patterns are equally appropriate as distance measures for predicting audibility thresholds. The Euclidean distance between specific loudness is worse in performance compared with the other two.

Speech formant frequency estimation: evaluating a nonstationary analysis method

Abstract The objective of this paper is to critically evaluate the performance of a nonstationary analysis method in tracking speech formant frequencies as they change with time due to the natural variations in the vocal-tract system during speech production. The method of instantaneous frequency estimation is applied to the tracking of speech formant frequencies to observe the time variations in the vocal-tract system characteristics within a pitch period. An implementation of an instantaneous frequency estimator based on the source–filter model of speech production is described for voiced speech formants. Based on experimental results from simulated as well as natural speech data, it is shown that the accuracy of the frequency estimates is heavily dependent on the nature of the glottal excitation waveform, the fundamental frequency and the frequency spacing of the formants in the speech signal. The choice of various analysis parameters on the accuracy of the estimates is discussed. It is shown that only when the formants are well separated and there are distinct regions of the glottal cycle in which the source excitation can be considered to be negligible, does the instantaneous frequency estimate accurately represent the actual formant frequency. Experimental results on natural speech vowels which show differences in formant frequencies in the different phases of the glottal cycle are presented.

An Overview of Hindustani Music in the Context of Computational Musicology

Abstract With its origin in the Samveda, composed between 1500–900 BC, the art music of India has evolved through ages and come to be regarded as one of the oldest surviving music systems in the world today. This paper aims to provide an overview of the fundamentals governing Hindustani music (also known as North Indian music) as practiced today. The deliberation will mainly focus on the melodic aspect of music making and will attempt to provide a musicological base for the main features associated with the melody: intonation and improvisation; thus covering the soundscape on the micro as well as macro level. The larger objective of this endeavour is to identify relevant directions for the application of computational approaches to Hindustani music.

A melody detection user interface for polyphonic music

The automatic extraction of the melody of the music from polyphonic recordings is a challenging problem for which no general solutions currently exist. We present a novel interface for semi-automatic melody extraction with the goal to provide highly accurate pitch tracks of the lead voice with minimal user intervention. Audio-visual feedback facilitates the validation of the obtained melodic contour and user control of the analysis parameters enables direct and effective control over the voice pitch detection module by the intelligent user. This paper describes the interface and discusses the features in which it differs from more conventional audio analyses interfaces.

Context-Aware features for singing voice detection in polyphonic music

The effectiveness of audio content analysis for music retrieval may be enhanced by the use of available metadata. In the present work, observed differences in singing style and instrumentation across genres are used to adapt acoustic features for the singing voice detection task. Timbral descriptors traditionally used to discriminate singing voice from accompanying instruments are complemented by new features representing the temporal dynamics of source pitch and timbre. A method to isolate the dominant source spectrum serves to increase the robustness of the extracted features in the context of polyphonic audio. While demonstrating the effectiveness of combining static and dynamic features, experiments on a culturally diverse music database clearly indicate the value of adapting feature sets to genre-specific acoustic characteristics. Thus commonly available metadata, such as genre, can be useful in the front-end of an MIR system.

Meter detection from audio for indian music

The meter of a musical excerpt provides high-level rhythmic information and is valuable in many music information retrieval tasks. We investigate the use of a computationally efficient approach to metrical analysis based on psycho-acoustically motivated decomposition of the audio signal. A two-stage comb filter-based approach, originally proposed for double/ triple meter estimation, is extended to a septuple meter (such as 7/8 time-signature) and its performance evaluated on a sizable Indian music database. We find that this system works well for Indian music and the distribution of musical stress/accents across a temporal grid can be utilized to obtain the metrical structure of audio automatically.