STUDENT'S-t MIXTURE MODEL BASED MULTI-INSTRUMENT RECOGNITION IN POLYPHONIC MUSIC

We address the problem of multi-instrument recognition in polyphonic music signals. Individual instruments are modeled within a stochastic framework using Student's-t Mixture Models (tMMs). We impose a mixture of these instrument models on the polyphonic signal model. No a priori knowledge is assumed about the number of instruments in the polyphony. The mixture weights are estimated in a latent variable framework from the polyphonic data using an Expectation Maximization (EM) algorithm, derived for the proposed approach. The weights are shown to indicate instrument activity. The output of the algorithm is an Instrument Activity Graph (IAG), using which, it is possible to find out the instruments that are active at a given time. An average F-ratio of 0 : 7 5 is obtained for polyphonies containing 2-5 instruments, on a experimental test set of 8 instruments: clarinet, flute, guitar, harp, mandolin, piano, trombone and violin.

Automatic Tonic Identification in Indian Art Music: Approaches and Evaluation

The tonic is a fundamental concept in Indian art music. It is the base pitch, which an artist chooses in order to construct the melodies during a rg(a) rendition, and all accompanying instruments are tuned using the tonic pitch. Consequently, tonic identification is a fundamental task for most computational analyses of Indian art music, such as intonation analysis, melodic motif analysis and rg recognition. In this paper we review existing approaches for tonic identification in Indian art music and evaluate them on six diverse datasets for a thorough comparison and analysis. We study the performance of each method in different contexts such as the presence/absence of additional metadata, the quality of audio data, the duration of audio data, music tradition (Hindustani/Carnatic) and the gender of the singer (male/female). We show that the approaches that combine multi-pitch analysis with machine learning provide the best performance in most cases (90% identification accuracy on average), and are robust across the aforementioned contexts compared to the approaches based on expert knowledge. In addition, we also show that the performance of the latter can be improved when additional metadata is available to further constrain the problem. Finally, we present a detailed error analysis of each method, providing further insights into the advantages and limitations of the methods.

EXPERIENCE OF OPERATING OPEN TOP BIOMASS GASIFIER USING A LOW DENSITY FUEL-COCONUT FRONDS

Use of fuel other than woody generally has been limited to rice husk and other residues are rarely tried as a fuel in a gasification system. With the availability of woody biomass in most countries like India, alternates fuels are being explored for sustainable supply of fuel. Use of agro residues has been explored after briquetting. There are few feedstock's like coconut fronts, maize cobs, etc, that might require lesser preprocessing steps compared to briquetting. The paper presents a detailed investigation into using coconut fronds as a fuel in an open top down draft gasification system. The fuel has ash content of 7% and was dried to moisture levels of 12 %. The average bulk density was found to be 230 kg/m3 with a fuel size particle of an average size 40 mm as compared to 350 kg/m3 for a standard wood pieces. A typical dry coconut fronds weighs about 2.5kgs and on an average 6 m long and 90 % of the frond is the petiole which is generally used as a fuel. The focus was also to compare the overall process with respect to operating with a typical woody biomass like subabul whose ash content is 1 %. The open top gasification system consists of a reactor, cooling and cleaning system along with water treatment. The performance parameters studied were the gas composition, tar and particulates in the clean gas, water quality and reactor pressure drop apart from other standard data collection of fuel flow rate, etc. The average gas composition was found to be CO 15 1.0 % H-2 16 +/- 1% CH4 0.5 +/- 0.1 % CO2 12.0 +/- 1.0 % and rest N2 compared to CO 19 +/- 1.0 % H-2 17 +/- 1.0 %, CH4 1 +/- 0.2 %, CO2 12 +/- 1.0 % and rest N2. The tar and particulate content in the clean gas has been found to be about 10 and 12 mg/m3 in both cases. The presence of high ash content material increased the pressure drop with coconut frond compared to woody biomass.

MULTI-INSTRUMENT DETECTION IN POLYPHONIC MUSIC USING GAUSSIAN MIXTURE BASED FACTORIAL HMM

We formulate the problem of detecting the constituent instruments in a polyphonic music piece as a joint decoding problem. From monophonic data, parametric Gaussian Mixture Hidden Markov Models (GM-HMM) are obtained for each instrument. We propose a method to use the above models in a factorial framework, termed as Factorial GM-HMM (F-GM-HMM). The states are jointly inferred to explain the evolution of each instrument in the mixture observation sequence. The dependencies are decoupled using variational inference technique. We show that the joint time evolution of all instruments' states can be captured using F-GM-HMM. We compare performance of proposed method with that of Student's-t mixture model (tMM) and GM-HMM in an existing latent variable framework. Experiments on two to five polyphony with 8 instrument models trained on the RWC dataset, tested on RWC and TRIOS datasets show that F-GM-HMM gives an advantage over the other considered models in segments containing co-occurring instruments.

Social Experience Is Sufficient to Modulate Sleep Need of Drosophila without Increasing Wakefulness

Organisms quickly learn about their surroundings and display synaptic plasticity which is thought to be critical for their survival. For example, fruit flies Drosophila melanogaster exposed to highly enriched social environment are found to show increased synaptic connections and a corresponding increase in sleep. Here we asked if social environment comprising a pair of same-sex individuals could enhance sleep in the participating individuals. To study this, we maintained individuals of D. melanogaster in same-sex pairs for a period of 1 to 4 days, and after separation, monitored sleep of the previously socialized and solitary individuals under similar conditions. Males maintained in pairs for 3 or more days were found to sleep significantly more during daytime and showed a tendency to fall asleep sooner as compared to solitary controls (both measures together are henceforth referred to as ``sleep-enhancement''). This sleep phenotype is not strain-specific as it is observed in males from three different ``wild type'' strains of D. melanogaster. Previous studies on social interaction mediated sleep-enhancement presumed `waking experience' during the interaction to be the primary underlying cause; however, we found sleep-enhancement to occur without any significant increase in wakefulness. Furthermore, while sleep-enhancement due to group-wise social interaction requires Pigment Dispersing Factor (PDF) positive neurons; PDF positive and CRYPTOCHROME (CRY) positive circadian clock neurons and the core circadian clock genes are not required for sleep-enhancement to occur when males interact in pairs. Pair-wise social interaction mediated sleep-enhancement requires dopamine and olfactory signaling, while visual and gustatory signaling systems seem to be dispensable. These results suggest that socialization alone (without any change in wakefulness) is sufficient to cause sleep-enhancement in fruit fly D. melanogaster males, and that its neuronal control is context-specific.