Computer Science - Sound Publications (50)


Computer Science - Sound Publications

To stretch a music piece to a given length is a common demand in people's daily lives, e.g., in audio-video synchronization and animation production. Read More

Behavioral annotation using signal processing and machine learning is highly dependent on training data and manual annotations of behavioral labels. Previous studies have shown that speech information encodes significant behavioral information and be used in a variety of automated behavior recognition tasks. However, extracting behavior information from speech is still a difficult task due to the sparseness of training data coupled with the complex, high-dimensionality of speech, and the complex and multiple information streams it encodes. Read More

In this paper, a novel architecture for a deep recurrent neural network, residual LSTM is introduced. A plain LSTM has an internal memory cell that can learn long term dependencies of sequential data. It also provides a temporal shortcut path to avoid vanishing or exploding gradients in the temporal domain. Read More

The higher order differential energy operator (DEO), denoted via $\Upsilon_k(x)$, is an extension to the second order famous Teager-Kaiser operator. The DEO helps measuring the higher order gauge of energy of a signal which is useful for AM-FM demodulation. However, the energy criterion defined by the DEO is not compliant with the presumption of positivity of energy. Read More

We propose a new deep network for audio event recognition, called AENet. In contrast to speech, sounds coming from audio events may be produced by a wide variety of sources. Furthermore, distinguishing them often requires analyzing an extended time period due to the lack of clear sub-word units that are present in speech. Read More

Speechreading is a notoriously difficult task for humans to perform. In this paper we present an end-to-end model based on a convolutional neural network (CNN) for generating an intelligible acoustic speech signal from silent video frames of a speaking person. The proposed CNN generates sound features for each frame based on its neighboring frames. Read More

Traditional speech enhancement techniques modify the magnitude of a speech in time-frequency domain, and use the phase of a noisy speech to resynthesize a time domain speech. This work proposes a complex-valued Gaussian process latent variable model (CGPLVM) to enhance directly the complex-valued noisy spectrum, modifying not only the magnitude but also the phase. The main idea that underlies the developed method is the modeling of short-time Fourier transform (STFT) coefficients across the time frames of a speech as a proper complex Gaussian process (GP) with noise added. Read More

There is a common observation that audio event classification is easier to deal with than detection. So far, this observation has been accepted as a fact and we lack a careful analysis. In this paper, we reason the rationale behind this fact and, more importantly, leverage them to benefit the audio event detection task. Read More

We introduce a dataset for facilitating audio-visual analysis of musical performances. The dataset comprises a number of simple multi-instrument musical pieces assembled from coordinated but separately recorded performances of individual tracks. For each piece, we provide the musical score in MIDI format, the audio recordings of the individual tracks, the audio and video recording of the assembled mixture, and ground-truth annotation files including frame-level and note-level transcriptions. Read More

In this paper we propose a novel model for unconditional audio generation based on generating one audio sample at a time. We show that our model, which profits from combining memory-less modules, namely autoregressive multilayer perceptrons, and stateful recurrent neural networks in a hierarchical structure is able to capture underlying sources of variations in the temporal sequences over very long time spans, on three datasets of different nature. Human evaluation on the generated samples indicate that our model is preferred over competing models. Read More

Most of the existing studies on voice conversion (VC) are conducted in acoustically matched conditions between source and target signal. However, the robustness of VC methods in presence of mismatch remains unknown. In this paper, we report a comparative analysis of different VC techniques under mismatched conditions. Read More

About 90 percent of people with Parkinson's disease (PD) experience decreased functional communication due to the presence of voice and speech disorders associated with dysarthria that can be characterized by monotony of pitch (or fundamental frequency), reduced loudness, irregular rate of speech, imprecise consonants, and changes in voice quality. Speech-language pathologists (SLPs) work with patients with PD to improve speech intelligibility using various intensive in-clinic speech treatments. SLPs also prescribe home exercises to enhance generalization of speech strategies outside of the treatment room. Read More

This paper addresses the problem of Target Activity Detection (TAD) for binaural listening devices. TAD denotes the problem of robustly detecting the activity of a target speaker in a harsh acoustic environment, which comprises interfering speakers and noise (cocktail party scenario). In previous work, it has been shown that employing a Feed-forward Neural Network (FNN) for detecting the target speaker activity is a promising approach to combine the advantage of different TAD features (used as network inputs). Read More

In this work, we propose a two-dimensional Head-Related Transfer Function (HRTF)-based robust beamformer design for robot audition, which allows for explicit control of the beamformer response for the entire three-dimensional sound field surrounding a humanoid robot. We evaluate the proposed method by means of both signal-independent and signal-dependent measures in a robot audition scenario. Our results confirm the effectiveness of the proposed two-dimensional HRTF-based beamformer design, compared to our previously published one-dimensional HRTF-based beamformer design, which was carried out for a fixed elevation angle only. Read More

We introduce a novel approach to studying animal behaviour and the context in which it occurs, through the use of microphone backpacks carried on the backs of individual free-flying birds. These sensors are increasingly used by animal behaviour researchers to study individual vocalisations of freely behaving animals, even in the field. However such devices may record more than an animals vocal behaviour, and have the potential to be used for investigating specific activities (movement) and context (background) within which vocalisations occur. Read More

This paper describes a computational model of loudness variations in expressive ensemble performance. The model predicts and explains the continuous variation of loudness as a function of information extracted automatically from the written score. Although such models have been proposed for expressive performance in solo instruments, this is (to the best of our knowledge) the first attempt to define a model for expressive performance in ensembles. Read More

In this paper, we describe three neural network (NN) based EEG-Speech (NES) models that map the unspoken EEG signals to the corresponding phonemes. Instead of using conventional feature extraction techniques, the proposed NES models rely on graphic learning to project both EEG and speech signals into deep representation feature spaces. This NN based linear projection helps to realize multimodal data fusion (i. Read More

We propose a new algorithm for time stretching music signals based on the theory of nonstationary Gabor frames. The algorithm extends the techniques of the classical phase vocoder by incorporating adaptive time-frequency representations and adaptive phase locking. Applying a preliminary onset detection algorithm, the obtained time-frequency representation implies good time resolution for the onsets and good frequency resolution for the sinusoidal components. Read More

This paper describes the LIA speaker recognition system developed for the Speaker Recognition Evaluation (SRE) campaign. Eight sub-systems are developed, all based on a state-of-the-art approach: i-vector/PLDA which represents the mainstream technique in text-independent speaker recognition. These sub-systems differ: on the acoustic feature extraction front-end (MFCC, PLP), at the i-vector extraction stage (UBM, DNN or two-feats posteriors) and finally on the data-shifting (IDVC, mean-shifting). Read More

In an attempt at exploring the limitations of simple approaches to the task of piano transcription (as usually defined in MIR), we conduct an in-depth analysis of neural network-based framewise transcription. We systematically compare different popular input representations for transcription systems to determine the ones most suitable for use with neural networks. Exploiting recent advances in training techniques and new regularizers, and taking into account hyper-parameter tuning, we show that it is possible, by simple bottom-up frame-wise processing, to obtain a piano transcriber that outperforms the current published state of the art on the publicly available MAPS dataset -- without any complex post-processing steps. Read More

Chord recognition systems depend on robust feature extraction pipelines. While these pipelines are traditionally hand-crafted, recent advances in end-to-end machine learning have begun to inspire researchers to explore data-driven methods for such tasks. In this paper, we present a chord recognition system that uses a fully convolutional deep auditory model for feature extraction. Read More

In this demo we show a novel approach to score following. Instead of relying on some symbolic representation, we are using a multi-modal convolutional neural network to match the incoming audio stream directly to sheet music images. This approach is in an early stage and should be seen as proof of concept. Read More

This paper demonstrates the feasibility of learning to retrieve short snippets of sheet music (images) when given a short query excerpt of music (audio) -- and vice versa --, without any symbolic representation of music or scores. This would be highly useful in many content-based musical retrieval scenarios. Our approach is based on Deep Canonical Correlation Analysis (DCCA) and learns correlated latent spaces allowing for cross-modality retrieval in both directions. Read More

We explore frame-level audio feature learning for chord recognition using artificial neural networks. We present the argument that chroma vectors potentially hold enough information to model harmonic content of audio for chord recognition, but that standard chroma extractors compute too noisy features. This leads us to propose a learned chroma feature extractor based on artificial neural networks. Read More

The use of deep learning to solve problems in literary arts has been a recent trend that has gained a lot of attention and automated generation of music has been an active area. This project deals with the generation of music using raw audio files in the frequency domain relying on various LSTM architectures. Fully connected and convolutional layers are used along with LSTM's to capture rich features in the frequency domain and increase the quality of music generated. Read More

Some glottal analysis approaches based upon linear prediction or complex cepstrum approaches have been proved to be effective to estimate glottal source from real speech utterances. We propose a new approach employing both an all-pole odd-order linear prediction to provide a coarse estimation and phase decomposition based causality/anti-causality separation to generate further refinements. The obtained measures show that this method improved performance in terms of reducing source-filter separation in estimation of glottal flow pulses (GFP). Read More

This paper proposed a class of novel Deep Recurrent Neural Networks which can incorporate language-level information into acoustic models. For simplicity, we named these networks Recurrent Deep Language Networks (RDLNs). Multiple variants of RDLNs were considered, including two kinds of context information, two methods to process the context, and two methods to incorporate the language-level information. Read More

We introduce a method for imposing higher-level structure on generated, polyphonic music. A Convolutional Restricted Boltzmann Machine (C-RBM) as a generative model is combined with gradient descent constraint optimization to provide further control over the generation process. Among other things, this allows for the use of a "template" piece, from which some structural properties can be extracted, and transferred as constraints to newly generated material. Read More

This paper introduces a new paradigm for sound source lo-calization referred to as virtual acoustic space traveling (VAST) and presents a first dataset designed for this purpose. Existing sound source localization methods are either based on an approximate physical model (physics-driven) or on a specific-purpose calibration set (data-driven). With VAST, the idea is to learn a mapping from audio features to desired audio properties using a massive dataset of simulated room impulse responses. Read More

This paper presented our work on applying Recurrent Deep Stacking Networks (RDSNs) to Robust Automatic Speech Recognition (ASR) tasks. In the paper, we also proposed a more efficient yet comparable substitute to RDSN, Bi- Pass Stacking Network (BPSN). The main idea of these two models is to add phoneme-level information into acoustic models, transforming an acoustic model to the combination of an acoustic model and a phoneme-level N-gram model. Read More

State-of-the-art i-vector based speaker verification relies on variants of Probabilistic Linear Discriminant Analysis (PLDA) for discriminant analysis. We are mainly motivated by the recent work of the joint bayesian (JB) method, which is originally proposed for discriminant analysis in face verification. We apply JB to speaker verification and make three contributions beyond of the original JB. Read More

In this paper, we investigate DCTNet for audio signal classification. Its output feature is related to Cohen's class of time-frequency distributions. We introduce the use of adaptive DCTNet (A-DCTNet) for audio signals feature extraction. Read More

Several methods exist for a computer to generate music based on data including Markov chains, recurrent neural networks, recombinancy, and grammars. We explore the use of unit selection and concatenation as a means of generating music using a procedure based on ranking, where, we consider a unit to be a variable length number of measures of music. We first examine whether a unit selection method, that is restricted to a finite size unit library, can be sufficient for encompassing a wide spectrum of music. Read More

A cost effective approach to remote monitoring of protected areas such as marine reserves and restricted naval waters is to use passive sonar to detect, classify, localize, and track marine vessel activity (including small boats and autonomous underwater vehicles). Cepstral analysis of underwater acoustic data enables the time delay between the direct path arrival and the first multipath arrival to be measured, which in turn enables estimation of the instantaneous range of the source (a small boat). However, this conventional method is limited to ranges where the Lloyd's mirror effect (interference pattern formed between the direct and first multipath arrivals) is discernible. Read More

Applying generic media-agnostic summarization to music allows for higher efficiency in automatic processing, storage, and communication of datasets while also alleviating copyright issues. This process has already been proven useful in the context of music genre classification. In this paper, we generalize conclusions from previous work by evaluating the impact of generic summarization in music from a probabilistic perspective and agnostic relative to certain tasks. Read More

Many people enjoy classical symphonic music. Its diverse instrumentation makes for a rich listening experience. This diversity adds to the conductor's expressive freedom to shape the sound according to their imagination. Read More

Heart diseases constitute a global health burden, and the problem is exacerbated by the error-prone nature of listening to and interpreting heart sounds. This motivates the development of automated classification to screen for abnormal heart sounds. Existing machine learning-based systems achieve accurate classification of heart sound recordings but rely on expert features that have not been thoroughly evaluated on noisy recordings. Read More

We present a new music dataset that can be used for several music analysis tasks. Our major goal is to go beyond the existing limitations of available music datasets, which are either the small size of datasets with raw audio tracks, the availability and legality of the music data, or the lack of meta-data for artists analysis or song ratings for recommender systems. Existing datasets such as GTZAN, TagATune, and Million Song suffer from the previous limitations. Read More

Musical frequencies in Just Intonation are comprised of rational numbers. The structure of rational numbers is determined by prime factorisations. Just Intonation frequencies can be split into two components. Read More

This paper introduces ALYSIA: Automated LYrical SongwrIting Application. ALYSIA is based on a machine learning model using Random Forests, and we discuss its success at pitch and rhythm prediction. Next, we show how ALYSIA was used to create original pop songs that were subsequently recorded and produced. Read More

The composition of polyphonic chorale music in the style of J.S Bach has represented a major challenge in automatic music composition over the last decades. The art of Bach chorales composition involves combining four-part harmony with characteristic rhythmic patterns and typical melodic movements to produce musical phrases which begin, evolve and end (cadences) in a harmonious way. Read More

In this paper we present FRIDA---an algorithm for estimating directions of arrival of multiple wideband sound sources. FRIDA combines multi-band information coherently and achieves state-of-the-art resolution at extremely low signal-to-noise ratios. It works for arbitrary array layouts, but unlike the various steered response power and subspace methods, it does not require a grid search. Read More

In North Indian Classical Music, raga forms the basic structure over which individual improvisations is performed by an artist based on his/her creativity. The Alap is the opening section of a typical Hindustani Music (HM) performance, where the raga is introduced and the paths of its development are revealed using all the notes used in that particular raga and allowed transitions between them with proper distribution over time. In India, corresponding to each raga, several emotional flavors are listed, namely erotic love, pathetic, devotional, comic, horrific, repugnant, heroic, fantastic, furious, peaceful. Read More

The works of Rabindranath Tagore have been sung by various artistes over generations spanning over almost 100 years. there are few songs which were popular in the early years and have been able to retain their popularity over the years while some others have faded away. In this study we look to find cues for the singing style of these songs which have kept them alive for all these years. Read More

We introduce a new large-scale music dataset, MusicNet, to serve as a source of supervision and evaluation of machine learning methods for music research. MusicNet consists of hundreds of freely-licensed classical music recordings by 10 composers, written for 11 instruments, together with instrument/note annotations resulting in over 1 million temporal labels on 34 hours of chamber music performances under various studio and microphone conditions. We define a multi-label classification task to predict notes in musical recordings, along with an evaluation protocol. Read More

This text offers a personal and very subjective view on the current situation of Music Information Research (MIR). Motivated by the desire to build systems with a somewhat deeper understanding of music than the ones we currently have, I try to sketch a number of challenges for the next decade of MIR research, grouped around six simple truths about music that are probably generally agreed on, but often ignored in everyday research. Read More

Designing appropriate features for acoustic event recognition tasks is an active field of research. Expressive features should both improve the performance of the tasks and also be interpret-able. Currently, heuristically designed features based on the domain knowledge requires tremendous effort in hand-crafting, while features extracted through deep network are difficult for human to interpret. Read More

One key step in audio signal processing is to transform the raw signal into representations that are efficient for encoding the original information. Traditionally, people transform the audio into spectral representations, as a function of frequency, amplitude and phase transformation. In this work, we take a purely data-driven approach to understand the temporal dynamics of audio at the raw signal level. Read More

This paper presents an efficient implementation of the Wavenet generation process called Fast Wavenet. Compared to a naive implementation that has complexity O(2^L) (L denotes the number of layers in the network), our proposed approach removes redundant convolution operations by caching previous calculations, thereby reducing the complexity to O(L) time. Timing experiments show significant advantages of our fast implementation over a naive one. Read More

Despite the overwhelming success of deep learning in various speech processing tasks, the problem of separating simultaneous speakers in a mixture remains challenging. Two major difficulties in such systems are the arbitrary source permutation and unknown number of sources in the mixture. We propose a novel deep learning framework for single channel speech separation by creating attractor points in high dimensional embedding space of the acoustic signals which pull together the time-frequency bins corresponding to each source. Read More