Sound Processing Techniques Research Articles

Machine Learning and Sound Processing in Vocal Disease Detection

We present in this paper some of the existing machine learning and sound processing techniques involved in the medical process and show how they can be applied in the context of a vocal disease detection task. The machine learning techniques are the usual, ranging from decision trees to neural networks, and when it comes to the sound processing approaches some scientists probably are familiar with them from the speech processing or music analysis areas. However, the techniques are adapted to the particularities of the medical field -- data collection, verification, etc. Through the concrete example of the vocal disease detection task we worked on, we made some interesting observations, both on the classification ability of the model and on the influence of the data.

Klasifikasi Rentang Usia Dan Gender Dengan Deteksi Suara Menggunakan Metode Deep Learning Algoritma Cnn (Convolutional Neural Network)

The research discusses the identification of human voices based on gender by utilizing the differences in voice characteristics between males and females. In addition to differences in vocal tract size, factors such as length, thickness, and stiffness of the vocal cords also play a role in producing the differences in fundamental frequency between the two genders. Fundamental frequency serves as an indicator used in acoustic analysis to classify gender based on voice. In the automatic classification of voices, sound processing techniques and machine learning are key in system development. Gender recognition methods based on voice involve acoustic analysis using voice features such as fundamental frequency, formants, duration, intensity, and intonation patterns. Diverse voice datasets containing recordings of both male and female voices are used to train gender recognition models. From the results of research from modeling using CNN on audio to get 92% accuracy and for testing results it is good enough in classifying.
 Keywords – Deep Learning, Voice Recognition, Audio Classification, CNN, Gender

MoMusic: A Motion-Driven Human-AI Collaborative Music Composition and Performing System

The significant development of artificial neural network architectures has facilitated the increasing adoption of automated music composition models over the past few years. However, most existing systems feature algorithmic generative structures based on hard code and predefined rules, generally excluding interactive or improvised behaviors. We propose a motion based music system, MoMusic, as a AI real time music generation system. MoMusic features a partially randomized harmonic sequencing model based on a probabilistic analysis of tonal chord progressions, mathematically abstracted through musical set theory. This model is presented against a dual dimension grid that produces resulting sounds through a posture recognition mechanism. A camera captures the users' fingers' movement and trajectories, creating coherent, partially improvised harmonic progressions. MoMusic integrates several timbrical registers, from traditional classical instruments such as the piano to a new ''human voice instrument'' created using a voice conversion technique. Our research demonstrates MoMusic's interactiveness, ability to inspire musicians, and ability to generate coherent musical material with various timbrical registers. MoMusic's capabilities could be easily expanded to incorporate different forms of posture controlled timbrical transformation, rhythmic transformation, dynamic transformation, or even digital sound processing techniques.

A Decision-Making Methodology Based on Expert Systems Applied to Machining Tools Condition Monitoring

The workers operating and supervising machining tools are often in charge of monitoring a high number of parameters of the machining process, and they usually make use of, among others, cutting sound signals, for following-up and assessing that process. The interpretation of those signals is closely related to the operational conditions of the machine and to the work environment itself, because such signals are sensitive to changes in the process’ input parameters. Additionally, they could be considered as a valid indicator for detecting working conditions that either negatively affect the tools’ lifespan, or might even put the machine operators themselves at risk. In light of those circumstances, this work deals with the proposal and conceptual development of a new methodology for monitoring the work conditions of machining tools, based on expert systems that incorporate a reinforcement strategy into their knowledge base. By means of the combination of sound-processing techniques, together with the use of fuzzy-logic inference engines and hierarchization methods based on vague fuzzy numbers, it will be possible to determine existing undesirable behaviors in the machining tools, thus reducing errors, accidents and harmful failures, with consequent savings in time and costs. Aiming to show the potential for the use of this methodology, a concept test has been developed, implemented in the form of a short case study. The results obtained, even if they require more extensive validation, suggest that the methodology would allow for improving the performance and operation of machining tools, as well as the ergonomic conditions of the workplace.

A tristimulus-formant model for automatic recognition of call types of laying hens

An essential objective of Precision Livestock Farming (PLF) is to use sensors that monitor bio-responses that contain important information on the health, well-being and productivity of farmed animals. In the literature, vocalisations of animals have been shown to contain information that can enable farmers to improve their animal husbandry practices. In this study, we focus on the vocalisation bio-responses of birds and specifically develop a sound recognition technique for continuous and automatic assessment of laying hen vocalisations. This study introduces a novel feature called the “tristimulus-formant” for the recognition of call types of laying hens (i.e., vocalisation types). Tristimulus is considered to be a timbre that is equivalent to the colour attributes of vision. Tristimulus measures the mixture of harmonics in a given sound, which grouped into 3 sections according to the relative weights of the harmonics in the signal. Experiments were designed in which calls from 11 Hy-Line brown hens were recorded in a cage-free setting (4303 vocalisations were labelled from 168 h of sound recordings). Then, sound processing techniques were used to extract the features of each call type and to classify the vocalisations using the LabVIEW® software. For feature extraction, we focused on extracting the Mel frequency cepstral coefficients (MFCCs) and tristimulus-formant (TF) features. Then, two different classifiers, the backpropagation neural network (BPNN) and Gaussian mixture model (GMM), were applied to recognise different call types. Finally, comparative trials were designed to test the different recognition models. The results show that the MFCCs-12+BPNN model (12 variables) had the highest average accuracy of 94.9 ± 1.6% but had the highest model training time (3201 ± 119 ms). At the same time, the MFCCs-3+TF+BPNN model had fewer feature dimensionalities (6 variables) and required less training time (2633 ± 54 ms) than the MFCCs-12+BPNN model and could classify well without compromising accuracy (91.4 ± 1.4%). Additionally, the BPNN classifier was better than the GMM classifier in recognising laying hens’ calls. The novel model can classify chicken sounds effectively at a low computational cost, giving it considerable potential for large data analysis and online monitoring systems.

Hearing, listening and deep neural networks in hearing aids

Hearing aids have undergone vast changes in the last 30 years from basic analog sound processing techniques, to advanced digital technology, to Deep Neural Networks (DNNs) “on-the-chip” providing real-time sound processing. In addition to making sounds audible, advanced hearing aids with DNN on-the-chip are better able to provide clearer understanding of speech in noise, improve recall, maintain interaural loudness and timing differences, and improve the wearer’s ability to selectively attend to the speaker of choice in challenging listening situations. These improvements are delivered without acoustic feedback and with very high sound quality.

A new approach to creating and deploying audio description for live theater

Audio description is an accessibility service used by blind or visually impaired individuals. Often accompanying movies, television shows, and other visual art forms, the service provides spoken descriptions of visual content, allowing people with vision loss the ability to access information that sighted people obtain visually. At live theatrical events, audio description provides spoken descriptions of scenes, characters, props, and other visual elements that those with vision loss may otherwise find inaccessible. This paper explores a novel approach to automate the creation and deployment of audio description for repeatable live theatrical events, both musical and non-musical. Using readily available tools and established sound-processing techniques, we describe a framework to automate several aspects of audio description for theater. The method uses a reference audio recording and an online time warping algorithm to align audio description with live performances, including a process for handling unexpected interruptions. We also show how a reference audio recording and descriptive tracks can be generated automatically from a show’s script and used to facilitate the deployment of audio description without sufficient resources for multiple performances or a live audio describer. Finally, a software implementation that is integrated into an existing theatrical workflow is also described. This system is used in three evaluation experiments that show the method successfully aligns multiple recordings of works of musical theater and non-musical plays in order to automatically trigger pre-recorded, descriptive audio in real time.

Intertext Variability of Smoothed Cepstral Peak Prominence, Methods to Control It, and Its Diagnostic Properties

This study aimed to estimate the intertext variability of smoothed cepstral peak prominence (CPPS), examine whether sound-processing techniques improved its variability and diagnostic capability, and evaluate the degree of intertext variability in detail with reference to the CPPS variabilities in sustained vowels. This was a retrospective study. Text readings of 58 Japanese syllables were recorded from 210 speakers with different diagnoses and varying degrees of dysphonia, and were divided into six passages. Applying the sound-processing techniques to those passages, we prepared three sample types: (1) nonprocessed, (2) only-loud, and (3) only-voiced samples. The intertext CPPS variability and diagnostic properties were compared across the passages and sample types. For detailed analysis, we subsequently extracted 63 normophonic speakers who maintained constant quality in their vowel utterances to evaluate the degree of intertext CPPS variability in relation to the variabilities between repeated identical vowels and across different vowels. Although several combinations of passages showed moderate-to-large CPPS variabilities, those variabilities were decreased by either technique, especially the deletion of silent segments, which resulted in the best diagnostic accuracy. The degree of intertext CPPS variability for the only-voiced samples was comparable to that of the CPPS variabilities in sustained vowels. The sound-processing technique removing silent segments should be applied to enhance the diagnostic properties of CPPS. The additional technique of deleting unvoiced segments is worth adopting if clinicians and researchers seek to attenuate the influence of text differences in calculating CPPS values.

Definition and Clinical Implication of Temporal Fine Structure

Perception of sound is associated with the use of multiple acoustic cues in the human auditory system. These acoustic cues can be classified by their temporal and spectral properties. Temporal information can be considered as a slowly varying envelope (ENV) superimposed on a more rapid temporal fine structure (TFS). Various sound-processing techniques have been developed to assess the role of acoustic information carried by temporal ENV and TFS information. Previous studies demonstrated that ENV cues are associated with speech recognition in quiet, while TFS cues have a role to melody/pitch perception and listening to speech in a competing background noise. Besides, a number of studies have shown that the relative insensitivity of hearing-impaired subjects to TFS information. Although current technologies used in cochlear implants are not efficient in delivering the TFS cues, new speech processing strategies have been proposed to deliver TFS information. We herein review the cutting edge of TFS related studies and discuss the clinical implication of TFS. Key words: Cochlear implant ㆍ Hearing impairment ㆍ Speech perception ㆍ Temporal envelope ㆍ Temporal fine structure

What is temporal fine structure and why is it important?

Complex sound like speech can be characterized as the sum of number of amplitude-modulated signals representing the outputs of an array of narrow frequency bands. Temporal information at the output of each band can be separated into temporal fine structure (TFS), the rapid oscillations close to the center frequency and temporal envelope (ENV), slower amplitude modulations superimposed on the TFS. TFS information can be carried in the pattern of phase locking to the stimulus waveform, while ENV by the changes in firing rate over time. The relative importance of temporal ENV and TFS information in understanding speech has been studied using various sound-processing techniques. A number of studies demonstrated that ENV cues are associated with speech recognition in quiet, while TFS cues are possibly linked to melody/pitch perception and listening to speech in a competing background. However, there are evidences that recovered ENV from TFS as well as TFS itself may be partially responsible for speech recognition. Current technologies used in cochlear implants (CI) are not efficient in delivering the TFS cues, and new attempts have been made to deliver TFS information into sound-processing strategy in CI. We herein discuss the current updated findings of TFS with a literature review.

Personalised Product Design Using Virtual Interactive Techniques

Use of Virtual Interactive Techniques for personalized product design is described in this paper. Usually products are designed and built by considering general usage patterns and Prototyping is used to mimic the static or working behaviour of an actual product before manufacturing the product. The user does not have any control on the design of the product. Personalized design postpones design to a later stage. It allows for personalized selection of individual components by the user. This is implemented by displaying the individual components over a physical model constructed using Cardboard or Thermocol in the actual size and shape of the original product. The components of the equipment or product such as screen, buttons etc. are then projected using a projector connected to the computer into the physical model. Users can interact with the prototype like the original working equipment and they can select, shape, position the individual components displayed on the interaction panel using simple hand gestures. Computer Vision techniques as well as sound processing techniques are used to detect and recognize the user gestures captured using a web camera and microphone.

Modelling Gesture Based Ubiquitous Applications

A cost effective, gesture based modelling technique called Virtual Interactive Prototyping (VIP) is described in this paper. Prototyping is implemented by projecting a virtual model of the equipment to be prototyped. Users can interact with the virtual model like the original working equipment. For capturing and tracking the user interactions with the model image and sound processing techniques are used. VIP is a flexible and interactive prototyping method that has much application in ubiquitous computing environments. Different commercial as well as socio-economic applications and extension to interactive advertising of VIP are also discussed.

Quasi-Synchronous and Pitch-Synchronous Granular Sound Processing with Stampede II

Granular sound-processing techniques allow effective and drastic modification of existing sounds, including independent alteration of the sounds' inherent pitch and time flow (Truax 1994). These modifications produce artifacts, which provide a rich source for artistic exploitation. On the other hand, the artifacts impose limits on the subtlety of transformation. For a number of years now, the speech-processing community has used efficient techniques for independent alteration of the pitch and time flow of utterances (Roucos and Wilgus 1985; Moulines and Charpentier 1990). Technically, these methods are closely related to classical granular methods used for musical purposes. Stampede II, a real-time program, unifies a number of these approaches in a common framework that allows for both subtle and dramatic alteration of both recorded and live sounds.