MIRAI: Multi-hierarchical, FS-Tree Based Music Information Retrieval System

In computer music, instrument recognition is a critical part of sound modeling. Pitch, timbre, loudness, duration, and spatialization are all components of musical sounds. All of these components play a significant part in determining the quality of the tonal sound. It is possible to alter the first four parameters, but timbre always poses a challenge [6]. It was inevitable that timbre would take center stage. Musical instruments are distinguished from one other by their distinct sound quality, independent of their pitch or volume. To distinguish between monophonic and polyphonic music recordings, this method might be used. In Musical Information Retrieval, classification plays one of the critical role. Monophonic instrument classification can be found in literature with quiet a substantial combinations of features and classifiers. Polyphonic instrument classification witnessed less references in the literature and is still an area to be explored specifically when it comes to Indian Classical domain. The present paper exactly focusses on this experimentation. Several Indian instruments were used to produce training data sets for the proposed approach’s evaluation purposes. Among the instruments utilized are the flute, harmonium, and sitar. Statistical and spectral factors are used to classify Indian musical instruments along with the Artificial Intelligence-based methods. Hybrid features from multiple domains that extract essential musical properties are extracted. Accuracy is demonstrated through an Indian Musical Instrument SVM and GMM classification. With monophonic sounds, SVM and Polyphonic produce an average accuracy of 89% and 91%. GMM outperforms SVM in monophonic recordings by a factor of 96.33 and polyphonic recordings by a factor of 93.33, according to the results of the studies. The future scope of this recognition framework can be an Artificial Intelligence System with a system linked with the Industrial Internet of Things (IIOT) framework to develop a standalone system or application which can be used for real- time classification of instruments.

A computationally efficient algorithm is proposed for modeling and coding the time-varying spectra of musical sounds. The aim is to encode individual data sets and not the statistical properties of the sounds. A given sequence of acoustic feature vectors is modeled by finding such a set of “states” (anchor points in the feature space) that the input data can be efficiently represented by interpolating between them. The achieved modeling accuracy for a database of musical sounds was approximately two times better than that of a conventional “vector quantization” model where the input data was k-means clustered and the input data vectors were then replaced by their corresponding cluster centroids. The computational complexity of the proposed algorithm as a function of the input sequence length T is O(TlogT).

Artificial Intelligence (AI) has revolutionized the field of music analysis by enabling advanced sound recognition and classification techniques. In the recent year, the music industry has had transformative evolution in recent years, significantly impacting user engagement, creativity, and technological innovation across various domains, including entertainment, education, and therapy. Musical instrument recognition is an emerging field within this landscape that could be used in applications such as automated music transcription and intelligent recommendation systems and adaptive music generation. Using Deep Learning (DL) alongside Artificial Intelligence has dramatically improved how we hear things as it has become a robust analytics tool for patterns in audio. The intricate signals from audio data together with overlapping frequencies and instrument diversity present significant challenges to accurate musical instrument prediction and classification. Traditional structured machine learning models cannot be applied successfully to dealing with complicated patterns, in contrast DL models possess superior designed system architectures and stronger feature extracting ability. Through the integration of these DNN with default layers with support vector machines, an instrument recognition framework is presented in this research, on publicly available dataset of diverse instruments with three second clips combined with Mel Spectrogram and its audio features. The standard measures are used for measuring performance of models such as accuracy, precision, recall and f1-score. The proposed model DNN achieves 98% classification precision over the SVM baseline with accuracy of 96% using musical instrument dataset with 24 classes. Using this research as the base of concept, it is shown that DL is better than current proposed methodologies at improving audio transformation processes, and it promises potential in improving the state of the art at musical instrument identification techniques that would yield useful results to intelligent music systems and AI audio analysis methodologies.

This paper reports the result of Musical instrument recognition using fractional fourier transform (FRFT) based features. The FRFT features are computed by replacing conventional Fourier transform in Mel Frequecny Cepstral coefficient ( MFCC) with FRFT. The result of the system using FRFT is compared with the result of the system using Mel Frequency Cepstral Coefficients (MFCC), Wavelet and Timbrel features with different machine learning algorithms. The experimentation is performed on isolated musical sounds of 19 musical instruments covering four different instrument families. The system using FRFT features outperforms over MFCC, Wavelet and Timbrel features with 91.84% recognition accuracy for individual instruments. The system is tested on benchmarked McGill University musical sound database. The experimental result shows that musical sound signals can be better represented using FRFT.

In this paper, a method for pitch independent musical instrument recognition using artificial neural networks is presented. Spectral features including FFT coefficients, harmonic envelopes and cepstral coefficients are used to represent the musical instrument sounds for classification. The effectiveness of these features are compared by testing the performance of ANNs trained with each feature. Multi-layer perceptrons are also compared with Time-delay neural networks. The testing and training sets both consist of fifteen note samples per musical instrument within the chromatic scale from C3 to C6. Both sets consist of nine instruments from the string, brass and woodwind families. Best results were achieved with cepstrum coefficients with a classification accuracy of 88 percent using a time-delay neural network, which is on par with recent results using several different features.

The research presented in this paper aims to automatically detect the singing voice in traditional Tunisian music, taking into account the main characteristics of the sound of the voice in this particular music style. This means creating the possibility to automatically identify instrumental and singing sounds. Therefore different methods for the automatic classification of sounds using supervised learning algorithms were compared and evaluated. The research is divided into four successive stages. First, the extraction of features vectors from the audio tracks (through calculation of the parameters of sound perception) followed by the selection and transformation process of relevant features for singing/instrumental discrimination. Then, using learning algorithms, the instrumental and vocal classes were modeled from a manually annotated database. Finally, the evaluation of the decision-making process (indexing) was applied on the test part of the database. The musical databases used for this study consists of extracts from the national sound archives of Centre of Mediterranean and Arabic Music (CMAM) and recordings made especially for this research. The possibility to index audio data (classify/segment) into vocal and instrumental recognition allows for the retrieval of content-based information of musical databases.

Supporting music has the power to represent the culture of a person or group at a particular time. He understood to have a basis for understanding, determining the ways and actions in each note produced. The color of the voice, the pitch and the musical work in each region have many differences. Because in essence, music is a work of art that is here is an aesthetic symbol of every different cultural community.Gamelan is a musical instrument that represents or depicts Javanese culture. Some Javanese gamelan in Surakarta, Yogyakarta and Cirebon, among others are Kyai Guntur Madu, Kyai Guntur Sari, Kyai Naga Wilaga and Gong Sekati. This orchestra has a pentatonic scale in the slendro scale system and the pelog scale system. Each gamelan has an honorary name as it deserves a sacred heirloom. slendro on the gamelan used as the research sample. While the independent variables are the frequency and color of the sound of each chord.

The aim of this paper is to present solutions related to identifying musical data. These are discussed mainly on the basis of experiments carried out at the Multimedia Systems Department, Gdansk University of Technology, Gdansk, Poland. The topics presented in this paper include automatic recognition of musical instruments and separation of duet sounds. The classification process is shown as a three-layer process consisting of pitch extraction, parametrization, and pattern recognition. These three stages are discussed on the basis of experimental examples. Artificial neural networks (ANNs) are employed as a decision system and they are trained with a set of feature vectors (FVs) extracted from musical sounds recorded at the Multimedia Systems Department. The frequency envelope distribution (FED) algorithm is presented, which was introduced to musical duet separation. For the purpose of checking the efficiency of the FED algorithm, ANNs are also used. They are tested on FVs derived from musical sounds after the separation process is performed. The experimental results are shown and discussed.

Introduction: The objective of our study was to evaluate musical perception and its relation to the quality of life in patients with bimodal binaural auditory stimulation.Materials and Methods: Nineteen adult patients with a cochlear implant (CI) for minimum 6 months, and moderate to severe contralateral hearing loss with a hearing aid (HA), and 21 normal hearing adults were included in this prospective, cross-sectional study. Pure-tone and speech audiometry, musical test evaluating sound perception characteristics and musical listening abilities, Munich questionnaire for musical habits, and the APHAB questionnaire were recoded. Performance in musical perception test with HA, CI, and HA + CI, and potential correlations between music test, audiometry and questionnaires were investigated.Results: Bimodal stimulation improved musical perception in several features (sound brightness, roughness, and clarity) in comparison to unimodal hearing, but CI did not add to HA performances in texture, polyphony or musical emotion and even appeared to interfere negatively in pitch perception with HA. Musical perception performances (sound clarity, instrument recognition) appeared to be correlated to hearing-related quality of life (APHAB RV and EC subdomains) but not with speech performances suggesting that the exploration of musical perception complements speech understanding evaluation to better describe every-day life hearing handicap.Conclusion: Testing musical sound perception provides important information on hearing performances as a complement to speech audiometry and appears to be related to hearing-related quality of life.

This paper describes a method that identifies sounds of non-registered musical instruments (i.e., musical instruments that are not contained in the training data) at a category level. Although the problem of how to deal with non-registered musical instruments is essential in musical instrument identification, it has not been dealt with in previous studies. Our method solves this problem by distinguishing between registered and non-registered instruments and identifying the category name of the non-registered instruments. When a given sound is registered, its instrument name, e.g. violin, is identified. Even if it is not registered, its category name, e.g. strings, can be identified. The important issue in achieving such identification is to adopt a musical instrument hierarchy reflecting the acoustical similarity. We present a method for acquiring such a hierarchy from a musical instrument sound database. Experimental results show that around 77% of non-registered instrument sounds, on average, were correctly identified at the category level.

Musical instruments radiate sound with specific directivity patterns in 3-dimensional space. The room transfer functions between a source and a receiver in concert halls also depend on the direction of sound radiation and arrival. Listeners perceive sound in 3-dimensional space. This paper explains the methods on the auralization of musical instruments in virtual concert halls that are measured or simulated. Musical instruments are played and recorded in anechoic chamber by equally quantizing the solid angle of radiation. Room transfer functions between a source and a receiver in halls are measured considering the direction of radiation and arrival. And then, the musical sounds considering the source directivity, direction of arrival, and all the room transfer functions are reproduced to give listeners the feeling that they are in the target hall. The methods are also applicable to live concert recording and purely virtual halls that are synthesized by room simulations.

Pressure waves, whether generated by humans or by natural processes, are a fundamental tool in the exploration of the Earth. The interaction between pressure waves and subsurface structures is complex, and diagnosing structure from observations of pressure waves is an active area of research. Many early studies of pressure waves and harmonic oscillations were motivated by a fascination with musical sounds, and musical instruments are still an excellent modeling system for exploring pressure-wave dynamics.

Objectives Cochlear implants (CI) have dramatically improved speech perception for patients with sensorineural hearing impairment. However, listening to music is a great challenge for them. This study examined the perception and appraisal of Iranian musical instruments comparing with similar Western instruments. Methods Eleven adult CI users and 25 normal hearing (NH) individuals participated in this study. Musical stimuli of three commonly heard instrument pairs were prepared. Participants were asked to identify the instruments and rate their appraisal on a ten-point Likert scale (0 = dislike very much, 10 = like very much). Results The instrument recognition rate was 40.6% among the CI users, and the mean appraisal score was 5.2 ± 2.7. NH listeners had none significant higher scores on both tasks with a recognition rate of 50.0% and the mean appraisal score of 6.9 ± 1.5. Iranian instruments were more recognized in both groups. Regarding their appraisal, the mean score for both types was almost equal in the NH group, while CI users more appraised Iranian instruments. Conclusion In addition to better recognition of Iranian instruments, they were particularly better appraised in the CI group. Iranian instruments provide suitable musical pieces for CI recipients that can be considered in rehabilitation programs.

For musical data analysis and retrieval, instrument recognition plays an important role. It helps in music search by instruments, makes music transcription easier and more accurate. There is need of extensive study on Indian musical instruments for various applications. In this paper, a system has been described for musical instrument recognition in monophonic audio signals where the single sound source is active at a time. The proposed system uses samples from three Indian classical instruments i.e, flute, harmonium, and sitar. The extracted feature set includes statistical, temporal and spectral, cepstral and linear predictive features. For the classification task, experimental results have been provided using a Gaussian mixture model (GMM). The system has shown recognition accuracy of 93.18% (average) for a combination of MFCC as a feature and GMM as a classifier.

Comprehensive dataset of Portuguese folk instruments for computer vision and heritage research