Anil M. Nagathil

Cepstral modulation ratio regression (CMRARE) parameters for audio signal analysis and classification

In this paper we propose a new set of parameters for audio signal analysis and classification. These parameters are regressions computed on the normalized modulation spectrum of high-resolution cepstral coefficients. The parameter set is scalable in its size and gives a compact representation of the modulation content of speech and other audio signals. These parameters as well as the regression approximation error are well suited for characterizing audio signals in a unified framework. In particular we use a set of eight parameters in a speech/music/noise classification task in which we achieve a classification accuracy which compares very well with other approaches including static and dynamic MFCCs.

Optimal signal reconstruction from a constant-Q spectrum

In contrast to other well-known techniques for spectral analysis such as the discrete Fourier transform or the wavelet transform the constant-Q transform (CQT) matches the center frequencies of its sub-band filters to the frequency scale of western music and accounts for the requirement of frequency-dependent bandwidths. However, it does not possess a strict mathematical inverse. Therefore, we derive an optimal reconstruction method to recover a signal from its CQT spectrum. The reconstruction problem is posed as a segmented overdetermined minimization problem where the number of frequency bins is larger than the number of reconstructed signal samples in each segment. The framework also enables the reduction of the input-output latency of the CQT. The method is evaluated for different types of music signals as well as for speech and Gaussian noise. For classical music a reconstruction quality of 89 dB signal-to-noise ratio is achieved.

Hierarchical audio classification using cepstral modulation ratio regressions based on Legendre polynomials

In this work we present a scalable feature set which is obtained by fitting orthogonal polynomials to the normalized modulation spectrum of cepstral coefficients and which can be easily adapted to different classification tasks. The performance of the feature set is investigated in a hierarchically structured audio signal classification experiment and compared with other approaches reported in the literature. For the root categories speech, music and noise a classification accuracy of 95% is achieved. Subclasses such as male and female speech or different noise types are classified with an accuracy of 95% and 85%, respectively. In a 10-category musical genre discrimination experiment the proposed features exhibit an accuracy of 61%.

Spectral complexity reduction of music signals for mitigating effects of cochlear hearing loss

In this paper we study reduced-rank approximations of music signals in the constant-Q spectral domain as a means to reduce effects stemming from cochlear hearing loss. The rationale behind computing reduced-rank approximations is that they allow to reduce the spectral complexity of a music signal. The method is motivated by studies with cochlear implant listeners which have shown that solo instrumental music or music remixed at higher signal-to-interference ratios are preferred over complex music ensembles or orchestras. For computing the reduced-rank approximations we investigate methods based on principal component analysis and partial least squares analysis, and compare them to source separation algorithms. The strategies, which are applied to music with a predominant leading voice, are compared in terms of their ability for mitigating effects of simulated reduced frequency selectivity and with respect to source signal distortions. Established instrumental measures and a newly developed measure indicate a considerable reduction of the auditory distortion resulting from cochlear hearing loss. Furthermore, a listening test reveals a significant preference for the reduced-rank approximations in terms of melody clarity and ease of listening.

Audio signal classification in reverberant environments based on fuzzy-clustered ad-hoc microphone arrays

Audio signal classification suffers from the mismatch of environmental conditions when training data is based on clean and anechoic signals and test data is distorted by reverberation and signals from other sources. In this contribution we analyze the classification performance for such a scenario with two concurrently active sources in a simulated reverberant environment. To obtain robust classification results, we exploit the spatial distribution of ad-hoc microphone arrays to capture the signals and extract cepstral features. Based on these features only, we use unsupervised fuzzy clustering to estimate clusters of microphones which are dominated by one of the sources. Finally, signal classification based on clean and anechoic training data is performed for each of the cluster. The probability of cluster membership for each microphone is provided by the fuzzy clustering algorithm and is used to compute a weighted average of the feature vectors. It is shown that the proposed method exceeds the performance of classification based on single microphones.

Classification of reverberant audio signals using clustered ad hoc distributed microphones

In a real world scenario, the automatic classification of audio signals constitutes a difficult problem. Often, reverberation and interfering sounds reduce the quality of a target source signal. This results in a mismatch between test and training data when a classifier is trained on clean and anechoic data. To classify disturbed signals more accurately we make use of the spatial distribution of microphones from ad hoc microphone arrays. In the proposed algorithm clusters of microphones that either are dominated by one of the sources in an acoustic scenario or contain mainly signal mixtures and reverberation are estimated in the audio feature domain. Information is shared within and in between these clusters to create one feature vector for each cluster to classify the source dominating this cluster. We evaluate the algorithm using simultaneously active sound sources and different ad hoc microphone arrays in simulated reverberant scenarios and multichannel recordings of an ad hoc microphone setup in a real environment. The cluster based classification accuracy is higher than the accuracy based on single microphone signals and allows for a robust classification of simultaneously active sources in reverberant environments. HighlightsAd hoc microphone arrays are used to classify audio sources in a reverberant environment.Clusters of microphones are estimated and used for information exchange in the feature domain.Cluster based processing allows for a classification of audio sources with a high accuracy.The evaluation is based on simulated and recorded reverberant audio data.

Performance of Specific vs. Generic Feature Sets in Polyphonic Music Instrument Recognition

Instrument identification in polyphonic audio recordings is a complex task which is beneficial for many music information retrieval applications. Due to the strong spectro-temporal differences between the sounds of existing instruments, different instrument-related features are required for building individual classification models. In our work we apply a multi-objective evolutionary feature selection paradigm to a large feature set minimizing both the classification error and the size of the used feature set. We compare two different feature selection methods. On the one hand we aim at building specific tradeoff feature sets which work best for the identification of a particular instrument. On the other hand we strive to design a generic feature set which on average performs comparably for all instrument classification tasks. The experiments show that the selected generic feature set approaches the performance of the selected instrument-specific feature sets, while a feature set specifically optimized for identifying a particular instrument yields degraded classification results if it is applied to other instruments.

Music complexity prediction for cochlear implant listeners based on a feature-based linear regression model

This paper presents a model for predicting music complexity as perceived by cochlear implant (CI) users. To this end, 10 CI users and 19 normal-hearing (NH) listeners rated 12 selected music pieces on a bipolar music complexity scale and 5 other perception-related scales. The results indicate statistically significant differences in the ratings between CI and NH listeners. In particular, the ratings among different scales were significantly correlated for CI users, which hints at a common, hidden scale. The median complexity ratings by CI listeners and features accounting for high-frequency energy, spectral center of gravity, spectral bandwidth, and roughness were used to train a linear principal component regression model for an average CI user. The model was evaluated by means of cross-validation and using an independent database of processed chamber music signals for which music preferences scores by CI users were available. The predictions indicate a clear linear relationship with the preference scores, confirming the negative correlation between music complexity and music preference for CI users found in previous studies. The proposed model is a first step toward an instrumental evaluation procedure in the emerging field of music processing for CIs.

A feature-based linear regression model for predicting perceptual ratings of music by cochlear implant listeners

While speech quality and intelligibility prediction methods for normal-hearing and hearing-impaired listeners have found a lot of attention as a cost-saving complement to listening tests, analogous procedures for music signals are still rare. In this paper a method is proposed for predicting perceptual ratings of music as obtained by cochlear implant (CI) listeners. For this purpose a listening test with CI listeners was conducted, who were asked to provide their ratings for music excerpts on different scales. It is shown that principal component regression (PCR) is a suitable tool to model and accurately predict the median ratings of the CI listeners using timbre and pitch related signal features as predictor variables. These features describe signal characteristics such as high-frequency energy, spectral bandwidth and roughness. The proposed prediction model is a first step towards an instrumental evaluation procedure for music processing algorithms in hearing devices.

Segmentation of music signals based on explained variance ratio for applications in spectral complexity reduction

Since natural acoustic signals like speech or music exhibit a highly varying temporal structure, signal enhancement and feature extraction algorithms benefit from segmentation procedures which take the underlying signal structure into account. In this paper we present a novel unsupervised segmentation procedure for music signals which relies on an explained variance criterion in the eigenspace of the constant-Q spectral domain. The procedure is used in the context of a spectral complexity reduction method which mitigates effects of cochlear hearing loss. It is compared to a segmentation based on equidistant boundaries. The results demonstrate that the proposed segmentation procedure gives an improvement in terms of signal-to-artefacts ratio in comparison to corresponding equidistant boundaries segmentation.