Voice Database Research Articles

Machine Learning Models With Hyperparameter Optimization for Voice Pathology Classification on Saarbrücken Voice Database.

Early diagnosis and referral are crucial in the treatment of voice disorders. Contemporary investigations have indicated the efficacy of voice pathology detection systems in significantly contributing to the evaluation of voice disorders, facilitating early diagnosis of such pathologies. These systems leverage machine learning methodologies, widely applied across diverse domains, and exhibit particular potential in the realm of voice pathology classification. However, machine learning models and performance metrics employed in these studies vary significantly, making it challenging to determine the optimal model for voice pathology classification. In this study, healthy and pathological voices were classified with state-of-the-art machine learning models, and the performance results of the models were compared. The voice samples employed in our research were sourced from the Saarbrücken Voice Database, a reputable German database. Feature extraction from voice signals was conducted using the Mel Frequency Cepstral Coefficients method. To assess and enhance the models' performance adequately, we employed hyperparameter optimization and implemented a 10-fold cross-validation approach. The outcomes revealed that the support vector machine model exhibited the highest accuracy, achieving 99.19% and 99.50% accuracies in the classification of male and female voice pathologies, respectively.

Harnessing machine learning in diagnosing complex hoarseness cases

PurposeTraditional vocal fold pathology recognition typically requires expertise of laryngologists and advanced instruments, primarily through direct visualization. This study aims to augment this conventional paradigm by introducing a parallel diagnostic procedure. Our objective is to harness a machine-learning algorithm designed to discern intricate patterns within patients' voice recordings to distinguish not only between healthy and hoarse voices but also among various specific disorders. Materials and methodsWe employed a machine-learning algorithm, utilizing transfer learning on the HuBERT model with Saarbruecken Voice Database samples. The study was conducted in two stages: a binary classifier distinguishes healthy and hoarse voices, while a subsequent multi-class classifier identifies specific voice disorders. Data from 2103 sessions, including over 25,000 components, representing diverse pathologies and healthy individuals, was analyzed. The models were trained, validated, and tested with a focus on robustness and accuracy in diagnosis. ResultsThe binary classifier achieved 82 % accuracy in distinguishing healthy from pathological voices. The multi-class algorithm which aims to identify specific laryngeal disorders obtained the highest accuracy (>93 %) for Laryngeal Dystonia. Noteworthy is the persistent challenge posed by Laryngeal Dystonia, a condition lacking a definitive diagnostic modality. ConclusionsOur findings demonstrate the feasibility of utilizing machine-learning algorithms to process voice samples, categorizing them into distinct pathologies. This approach holds potential for enhance patient triage, streamline diagnostics, and elevate overall patient care. Particularly valuable for challenging diagnoses, such as Laryngeal Dystonia, this method underscores the transformative role of machine learning in optimizing healthcare practices.

Optimized early fusion of handcrafted and deep learning descriptors for voice pathology detection and classification

This study presents an automated noninvasive voice disorder detection and classification approach using an optimized fusion of modified glottal source estimation and deep transfer learning neural network descriptors. A new set of modified descriptors based on a glottal source estimator and pre-trained Inception-ResNet-v2 convolutional neural network-based features are proposed for the speech disorder detection and classification task. The modified feature set is obtained using mel-cepstral coefficients, harmonic model, phase discrimination means, distortion deviation descriptors, conventional wavelet, and glottal source estimation features. Early descriptor-level fusion is employed in this study for performance enhancement-however, the fusion results in higher feature vector dimensionality. A nature-inspired slime mould algorithm is utilized to remove redundant and select the best discriminating features. Finally, the classification is performed using the K-nearest neighbor (KNN) classifier. The proposed algorithm was evaluated using extensive experiments with different feature combinations, with and without feature selection, and with two popular datasets: the Arabic Voice Pathology Database (AVPD) and the Saarbrucken Voice Database (SVD). We show that the proposed optimized fusion method attained an enhanced voice pathology detection accuracy of 98.46%, encompassing a wide spectrum of voice disorders on the SVD database. Furthermore, compared to traditional handcrafted and deep neural network-based techniques, the proposed method demonstrates competitive performance with fewer features.

How well do acoustic parameters correlate our perception of voice quality?

Auditory perceptual judgments are essential for assessing voice disorders and monitoring treatment progress. This study aimed to examine the acoustic foundations of perceived abnormalities in individuals with and without voice disorders, focusing on the correlation between the Grade, Roughness, Breathiness, Asthenia, Strain (GRBAS) scale and various acoustic parameters. A total of 1217 voice recordings were obtained from the Perceptual Voice Qualities Database (PVQD) and the Saarbruecken Voice Database (SVD). Each recording was perceptually evaluated by five experienced speech therapists based on the GRBAS scale. Only ratings of more than 80% inter-rater agreement was used for further analysis. Acoustic analyses were obtained using Parselmouth, a Python library for the Praat software. Statistical analysis using Spearman’s rank correlation revealed strong correlations between spectral and cepstral results and GRBAS scale. Specifically, the Grade and Breathiness components showed significant negative correlations with the CPPS parameters, with and without voice detection. There was a strong inverse relationship between Harmonics-to-Noise Ratio (HNR) and the Grade and Roughness. Jitter and shimmer demonstrated robust positive correlations with Grade, Breathiness, and Roughness. These findings provide valuable insights into the relationships between acoustic parameters and perceptual GRBAS scale, informing future research and clinical practice in voice and speech pathology.

Multifeature Fusion Method with Metaheuristic Optimization for Automated Voice Pathology Detection

Voice pathologies occur due to various factors, such as malfunction of the vocal cords. Computerized acoustic examination-based vocal pathology detection is crucial for early diagnosis, efficient follow-up, and improving problematic speech. Different acoustic measurements provide it. Executing this process requires expert monitoring and is not preferred by patients because it is time-consuming and costly. This paper is aimed at detecting metaheuristic-based automatic voice pathology. First, feature maps of ten common diseases, including cordectomy, dysphonia, front lateral partial resection, contact pachyderma, laryngitis, lukoplakia, pure breath, recurrent laryngeal paralysis, vocal fold polyp, and vox senilis, were obtained from the Zero-Crossing Rate (ZCR), Root-Mean-Square Energy (RMSE), and Mel-frequency Cepstral Coefficients (MFCCs) using a thousand voice signals from the Saarbruecken Voice Database (SVD) dataset. Hybridizations of different features obtained from the voices of the same diseases using these three methods were used to increase the model's performance. The Grey Wolf Optimizer (MELGWO) algorithm based on local search, evolutionary operator, and concatenated feature maps derived from various approaches was employed to minimize the number of features, implement the models faster, and produce the best result. The fitness values of the metaheuristic algorithms were then determined using supervised machine learning techniques such as Support Vector Machine (SVM) and K-nearest neighbors (KNN). The F1 score, sensitivity, specificity, accuracy, and other assessment criteria were compared with the experimental data. The best accuracy result was achieved with 99.50% from the SVM classifier using the feature maps optimized by the improved MELGWO algorithms.

Tracking age-related changes in voice and speech production with Landmark-based analysis of speech.

Voice and speech production change with age, which can lead to potential communication challenges. This study explored the use of Landmark-based analysis of speech (LMBAS), a knowledge-based speech analysis algorithm based on Stevens' Landmark Theory, to describe age-related changes in adult speakers. The speech samples analyzed were sourced from the University of Florida Aging Voice Database, which included recordings of 16 sentences from the Speech Perception in Noise test of Bilger, Rzcezkowski, Nuetzel, and Rabinowitz [J. Acoust. Soc. Am. 65, S98-S98 (1979)] and Bilger, Nuetzel, Rabinowitz, and Rzeczkowski [J. Speech. Lang. Hear. Res. 27, 32-84 (1984)]. These sentences were read in quiet environments by 50 young, 50 middle-aged, and 50 older American English speakers, with an equal distribution of sexes. Acoustic landmarks, specifically, glottal, bursts, and syllabicity landmarks, were extracted using SpeechMark®, MATLAB Toolbox version 1.1.2. The results showed significant age effect on glottal and burst landmarks. Furthermore, the sex effect was significant for burst and syllabicity landmarks. While the results of LMBAS suggest its potential in detecting age-related changes in speech, increase in syllabicity landmarks with age was unexpected. This finding may suggest the need for further refinement and adjustment of this analytical approach.

Pathological voice classification using MEEL features and SVM-TabNet model

In clinical settings, early diagnosis and objective assessment depend on the detection of voice pathology. To classify anomalous voices, this work uses an approach that combines the SVM-TabNet fusion model with MEEL (Mel-Frequency Energy Line) features. Further, the dataset consists of 1037 speech files, including recordings from people with laryngocele and Vox senilis as well as from healthy persons. Additionally, the main goal is to create an efficient classification model that can differentiate between normal and abnormal voice patterns. Modern techniques frequently lack the accuracy required for a precise diagnosis, which highlights the need for novel strategies. The suggested approach uses an SVM-TabNet fusion model for classification after feature extraction using MEEL characteristics. MEEL features provide extensive information for categorization by capturing complex patterns in audio transmissions. Moreover, by combining the advantages of SVM and TabNet models, classification performance is improved. Moreover, testing the model on test data yields remarkable results: 99.7 % accuracy, 0.992 F1 score, 0.996 precision, and 0.995 recall. Additional testing on additional datasets reliably validates outstanding performance, with 99.4 % accuracy, 0.99 F1 score, 0.998 precision, and 0.989 % recall. Furthermore, using the Saarbruecken Voice Database (SVD), the suggested methodology achieves an impressive accuracy of 99.97 %, demonstrating its durability and generalizability across many datasets. Overall, this work shows how the SVM-TabNet fusion model with MEEL characteristics may be used to accurately and consistently classify diseased voices, providing encouraging opportunities for clinical diagnosis and therapy tracking.

Using Vocal-Based Emotions as a Human Error Prevention System with Convolutional Neural Networks

Human error is a mark assigned to an event that has negative effects or does not produce a desired result, with emotions playing an important role in how humans think and behave. If we detect feelings early, it may decrease human error. The human voice is one of the most powerful tools that can be used for emotion recognition. This study aims to reduce human error by building a system that detects positive or negative emotions of a user like (happy, sad, fear, and anger) through the analysis of the proposed vocal emotion component using Convolutional Neural Networks. By applying the proposed method to an emotional voice database (RAVDESS) using Librosa for voice processing and PyTorch, with the emotion classification of (happy/angry), the results show a better accuracy (98%) in comparison to the literature with regard to making a decision to deny or allow a user to access sensitive operations or send a warning to the system administrator prior to accessing system resources.

AROA based Pre-trained Model of Convolutional Neural Network for Voice Pathology Detection and Classification

With the demand for better, more user-friendly HMIs, voice recognition systems have risen in prominence in recent years. The use of computer-assisted vocal pathology categorization tools allows for the accurate detection of voice pathology diseases. By using these methods, vocal disorders may be diagnosed early on and treated accordingly. An effective Deep Learning-based tool for feature extraction-based vocal pathology identification is the goal of this project. This research presents the results of using EfficientNet, a pre-trained Convolutional Neural Network (CNN), on a speech pathology dataset in order to achieve the highest possible classification accuracy. An Artificial Rabbit Optimization Algorithm (AROA)-tuned set of parameters complements the model's mobNet building elements, which include a linear stack of divisible convolution and max-pooling layers activated by Swish. In order to make the suggested approach applicable to a broad variety of voice disorder problems, this study also suggests a unique training method along with several training methodologies. One speech database, the Saarbrücken voice database (SVD), has been used to test the proposed technology. Using up to 96% accuracy, the experimental findings demonstrate that the suggested CNN approach is capable of detecting speech pathologies. The suggested method demonstrates great potential for use in real-world clinical settings, where it may provide accurate classifications in as little as three seconds and expedite automated diagnosis and treatment.

Automated Acoustic Evaluation of Voice Disorders: A Comprehensive Study on Parameter Analysis Using ANN

Voice analysis serves as a crucial tool in diagnosing voice abnormalities, offering a non-invasive alternative to intrusive procedures. This research digs into a comprehensive study of voice disorder assessment methodologies, focusing on acoustic analysis and classification. The study employs various parameters such as Jitter, Shimmer, and Harmonic-to-Noise Ratio (HNR) alongside an Artificial Neural Network (ANN) classifier. Utilizing the Saarbruecken Voice Database the research aims to distinguish between healthy and dysphonic voices across genders. Principal Component Analysis (PCA) aids in feature selection, enhancing model accuracy. The results exhibit distinct precision levels in male and female groups, showcasing the effectiveness of specific parameters in classification.

Diagnosis of pathological speech with streamlined features for long short-term memory learning

Background:Pathological speech diagnosis is crucial for identifying and treating various speech disorders. Accurate diagnosis aids in developing targeted intervention strategies, improving patients’ communication abilities, and enhancing their overall quality of life. With the rising incidence of speech-related conditions globally, including oral health, the need for efficient and reliable diagnostic tools has become paramount, emphasizing the significance of advanced research in this field. Methods:This paper introduces novel features for deep learning in the analysis of short voice signals. It proposes the incorporation of time-space and time–frequency features to accurately discern between two distinct groups: Individuals exhibiting normal vocal patterns and those manifesting pathological voice conditions. These advancements aim to enhance the precision and reliability of diagnostic procedures, paving the way for more targeted treatment approaches. Results:Utilizing a publicly available voice database, this study carried out training and validation using long short-term memory (LSTM) networks learning on the combined features, along with a data balancing strategy. The proposed approach yielded promising performance metrics: 90% accuracy, 93% sensitivity, 87% specificity, 88% precision, an F1 score of 0.90, and an area under the receiver operating characteristic curve of 0.96. The results surpassed those obtained by the networks trained using wavelet-time scattering coefficients, as well as several algorithms trained with alternative feature types. Conclusions:The incorporation of time–frequency and time-space features extracted from short segments of voice signals for LSTM learning demonstrates significant promise as an AI tool for the diagnosis of speech pathology. The proposed approach has the potential to enhance the accuracy and allow for real-time pathological speech assessment, thereby facilitating more targeted and effective therapeutic interventions.

Vocal Tract Acoustic Measurements for Detection of Pathological Voice Disorders

Human voice is an important signal that, if analyzed cautiously, will reveal various disorders and diseases of the human body. Vocal profiling is emerging as a new technique for assessment of voice disorders. This work has focused mainly on vocal tract acoustic measurements to detect voice disorders, especially dysphonia. Dysphonia is a communication disorder that, if not detected at an early stage, may lead to serious complications like laryngeal cancer thus affecting health and quality of life. The voice recordings of 52 subjects (26 dysphonic and 26 healthy) were taken from Saarbrucken Voice Database, 28 live samples (14 dysphonic and 14 healthy) were recorded using CSL 4500 tool and 169 subjects (111 dysphonic and 58 healthy) were taken from VOICED Database. These voice samples were used to extract five acoustic parameters: fundamental frequency, jitter, shimmer, pitch, and formants. The results obtained had an accuracy of around 85% which confirms the potentiality of these fundamental parameters in the successful detection of dysphonia as well as other voice disorders.

Validation of an AI-assisted Treatment Outcome Measure for Gender-Affirming Voice Care: Comparing AI Accuracy to Listener’s Perception of Voice Femininity

OBJECTIVESThere is currently a lack of objective treatment outcome measures for transgender individuals undergoing gender-affirming voice care. Recently, Bensoussan et al. developed an AI model that is able to generate a voice femininity rating based on a short voice sample provided through a smartphone application. The purpose of this study was to examine the feasibility of using this model as a treatment outcome measure by comparing its performance to human listeners. Additionally, we examined the effect of two different training datasets on the model’s accuracy and performance when presented with external data. METHODS100 voice recordings from 50 cisgender males and 50 cisgender females were retrospectively collected from patients presenting at a university voice clinic for reasons other than dysphonia. The recordings were evaluated by expert and naïve human listeners, who rated each voice based on how sure they were the voice belonged to a female speaker (% voice femininity [R]). Human ratings were compared to ratings generated by (1) the AI model trained on a high-quality low-quantity dataset (voices from the Perceptual Voice Quality Database) (PVQD model), and (2) the AI model trained on a low-quality high-quantity dataset (voices from the Mozilla Common Voice database) (Mozilla model). Ambiguity scores were calculated as the absolute value of the difference between the rating and certainty (0 or 100%). RESULTSBoth expert and naïve listeners achieved 100% accuracy in identifying voice gender based on a binary classification (female >50% voice femininity [R]). In comparison, the Mozilla-trained model achieved 92% accuracy and the previously published PVQD model achieved 84% accuracy in determining voice gender (female > 50% AI voice femininity). While both AI models correlated with human ratings, the Mozilla-trained model showed a stronger correlation as well as lower overall rating ambiguity than the PVQD-trained model. The Mozilla model also appeared to handle pitch information in a similar way to human raters. CONCLUSIONSThe AI model predicted voice gender with high accuracy when compared to human listeners and has potential as a useful outcome measure for transgender individuals receiving gender-affirming voice training. The Mozilla-trained model performed better than the PVQD-trained model, indicating that for binary classification tasks, quantity of data may influence accuracy more than quality of the data used for training the voice AI models.

PVGAN: A Pathological Voice Generation Model Incorporating a Progressive Nesting Strategy

The voice generation task is to solve the problem of limited samples in the voice dataset using computer technology. By increasing the number of samples, the accuracy of voice disorder diagnosis can be improved, which has a wide range of application value in medical diagnosis and other fields. At present, there are insufficient models for detailed features such as pitch, timbre, and different frequency components in pathological voice data. Therefore, this paper proposes a PVGAN network for learning different frequency information of audio to generate pathological voice data. The proposed network captures the multi-scale features and different periodic patterns of audio signals by designing multiscale perceptual residual blocks and periodic discriminators. At the same time, a progressive nesting strategy was proposed to combine the generator and the discriminator to improve the learning ability of different resolution information. In addition, a latent mapping network is designed to fuse the latent vector with the condition information to generate sound features related to specific diseases or pathological states. The loss function is optimized to further improve the model performance. On the Saarbruecken Voice Database(SVD), the average values of each index of the data generated after training with different pathological types as conditional information are similar to the original data. Finally, the generated data were used to expand the SVD dataset, and the accuracy of the two classification experiments was improved to a certain extent.

Unraveling the complexities of pathological voice through saliency analysis

The human voice is an essential communication tool, but various disorders and habits can disrupt it. Diagnosis of pathological and abnormal voices is very important. Conventional diagnosis of these voice pathologies can be invasive and costly. Voice pathology disorders can be effectively detected using Artificial Intelligence and computer-aided voice pathology classification tools. Previous studies focused primarily on binary classification, leaving limited attention to multi-class classification. This study proposes three different neural network architectures to investigate the feature characteristics of three voice pathologies-Hyperkinetic Dysphonia, Hypokinetic Dysphonia, Reflux Laryngitis, and healthy voices using multi-class classification and the Voice ICar fEDerico II (VOICED) dataset. The study proposes UNet++ autoencoder-based denoiser techniques for accurate feature extraction to overcome noisy data. The architectures include a Multi-Layer Perceptron (MLP) trained on structured feature sets, a Short-Time Fourier Transform (STFT) model, and a Mel-Frequency Cepstral Coefficients (MFCC) model. The MLP model on 143 features achieved 97.1% accuracy, while the STFT model showed similar performance with increased sensitivity of 99.8%. The MFCC model maintained 97.1% accuracy but with a smaller model size and improved accuracy on the Reflux Laryngitis class. The study identifies crucial features through saliency analysis and reveals that detecting voice abnormalities requires the identification of regions of inaudible high-pitch sounds. Additionally, the study highlights the challenges posed by limited and disjointed pathological voice databases and proposes solutions for enhancing the performance of voice abnormality classification. Overall, the study’s findings have potential applications in clinical applications and specialized audio-capturing tools.

Validation of scrambling methods for vocal affect bursts

Studies on perception and cognition require sound methods allowing us to disentangle the basic sensory processing of physical stimulus properties from the cognitive processing of stimulus meaning. Similar to the scrambling of images, the scrambling of auditory signals is aimed at creating stimulus instances that are unrecognizable but have comparable low-level features. In the present study, we generated scrambled stimuli of short vocalizations taken from the Montreal Affective Voices database (Belin et al., Behav Res Methods, 40(2):531–539, 2008) by applying four different scrambling methods (frequency-, phase-, and two time-scrambling transformations). The original stimuli and their scrambled versions were judged by 60 participants for the apparency of a human voice, gender, and valence of the expressions, or, if no human voice was detected, for the valence of the subjective response to the stimulus. The human-likeness ratings were reduced for all scrambled versions relative to the original stimuli, albeit to a lesser extent for phase-scrambled versions of neutral bursts. For phase-scrambled neutral bursts, valence ratings were equivalent to those of the original neutral burst. All other scrambled versions were rated as slightly unpleasant, indicating that they should be used with caution due to their potential aversiveness.

End-to-end deep learning classification of vocal pathology using stacked vowels.

Advances in artificial intelligence (AI) technology have increased the feasibility of classifying voice disorders using voice recordings as a screening tool. This work develops upon previous models that take in single vowel recordings by analyzing multiple vowel recordings simultaneously to enhance prediction of vocal pathology. Voice samples from the Saarbruecken Voice Database, including three sustained vowels (/a/, /i/, /u/) from 687 healthy human participants and 334 dysphonic patients, were used to train 1-dimensional convolutional neural network models for multiclass classification of healthy, hyperfunctional dysphonia, and laryngitis voice recordings. Three models were trained: (1) a baseline model that analyzed individual vowels in isolation, (2) a stacked vowel model that analyzed three vowels (/a/, /i/, /u/) in the neutral pitch simultaneously, and (3) a stacked pitch model that analyzed the /a/ vowel in three pitches (low, neutral, and high) simultaneously. For multiclass classification of healthy, hyperfunctional dysphonia, and laryngitis voice recordings, the stacked vowel model demonstrated higher performance compared with the baseline and stacked pitch models (F1 score 0.81 vs. 0.77 and 0.78, respectively). Specifically, the stacked vowel model achieved higher performance for class-specific classification of hyperfunctional dysphonia voice samples compared with the baseline and stacked pitch models (F1 score 0.56 vs. 0.49 and 0.50, respectively). This study demonstrates the feasibility and potential of analyzing multiple sustained vowel recordings simultaneously to improve AI-driven screening and classification of vocal pathology. The stacked vowel model architecture in particular offers promise to enhance such an approach. AI analysis of multiple vowel recordings can improve classification of voice pathologies compared with models using a single sustained vowel and offer a strategy to enhance AI-driven screening of voice disorders. 3.

Automated voice pathology discrimination from audio recordings benefits from phonetic analysis of continuous speech

In this paper we evaluate the hypothesis that automated methods for diagnosis of voice disorders from speech recordings would benefit from contextual information found in continuous speech. Rather than basing a diagnosis on how disorders affect the average acoustic properties of the speech signal, the idea is to exploit the possibility that different disorders will cause different acoustic changes within different phonetic contexts. Any differences in the pattern of effects across contexts would then provide additional information for discrimination of pathologies. We evaluate this approach using two complementary studies: the first uses a short phrase which is automatically annotated using a phonetic transcription, the second uses a long reading passage which is automatically annotated from text. The first study uses a single sentence recorded from 597 speakers in the Saarbrucken Voice Database to discriminate structural from neurogenic disorders. The results show that discrimination performance for these broad pathology classes improves from 59% to 67% unweighted average recall when classifiers are trained for each phone-label and the results fused. Although the phonetic contexts improved discrimination, the overall sensitivity and specificity of the method seems insufficient for clinical application. We hypothesise that this is because of the limited contexts in the speech audio and the heterogeneous nature of the disorders. In the second study we address these issues by processing recordings of a long reading passage obtained from clinical recordings of 60 speakers with either Spasmodic Dysphonia or Vocal fold Paralysis. We show that discrimination performance increases from 80% to 87% unweighted average recall if classifiers are trained for each phone-labelled region and predictions fused. We also show that the sensitivity and specificity of a diagnostic test with this performance is similar to other diagnostic procedures in clinical use. In conclusion, the studies confirm that the exploitation of contextual differences in the way disorders affect speech improves automated diagnostic performance, and that automated methods for phonetic annotation of reading passages are robust enough to extract useful diagnostic information.

A Smart Image Encryption Technology via Applying Personal Information and Speaker-Verification System

In this paper, a framework for authorization and personal image protection that applies user accounts, passwords, and personal I-vectors as the keys for ciphering the image content was developed and connected. There were two main systems in this framework. The first involved a speaker verification system, wherein the user entered their account information and password to log into the system and provided a short voice sample for identification, and then the algorithm transferred the user’s voice (biometric) features, along with their account and password details, to a second image encryption system. For the image encryption process, the account name and password presented by the user were applied to produce the initial conditions for hyper-chaotic systems to generate private keys for image-shuffling and ciphering. In the final stage, the biometric features were also applied to protect the content of the image, so the encryption technology would be more robust. The final results of the encryption system were acceptable, as a lower correlation was obtained in the cipher images. The voice database we applied was the Pitch Tracking Database from the Graz University of Technology (PTDB-TUG), which provided the microphone and laryngoscope signals of 20 native English speakers. For image processing, four standard testing images from the University of Southern California–Signal and Image Processing Institute (USC-SIPI), including Lena, F-16, Mandrill, and Peppers, were presented to further demonstrate the effectiveness and efficiency of the smart image encryption algorithm.

Multiple voice disorders in the same individual: Investigating handcrafted features, multi-label classification algorithms, and base-learners

Non-invasive acoustic analyses of voice disorders have been at the forefront of current biomedical research. Usual strategies, essentially based on machine learning (ML) algorithms, commonly classify a subject as being either healthy or pathologically-affected. Nevertheless, the latter state is not always a result of a sole laryngeal issue, i.e., multiple disorders might exist, demanding multi-label classification procedures for effective diagnoses. Consequently, the objective of this paper is to investigate the application of five multi-label classification methods based on problem transformation to play the role of base-learners, i.e., Label Powerset, Binary Relevance, Nested Stacking, Classifier Chains, and Dependent Binary Relevance with Random Forest (RF) and Support Vector Machine (SVM), in addition to a Deep Neural Network (DNN) from an algorithm adaptation method, to detect multiple voice disorders, i.e., Dysphonia, Laryngitis, Reinke’s Edema, Vox Senilis, and Central Laryngeal Motion Disorder. Receiving as input three handcrafted features, i.e., signal energy (SE), zero-crossing rates (ZCRs), and signal entropy (SH), which allow for interpretable descriptors in terms of speech analysis, production, and perception, we observed that the DNN-based approach powered with SE-based feature vectors presented the best values of F1-score among the tested methods, i.e., 0.943, as the averaged value from all the balancing scenarios, under Saarbrücken Voice Database (SVD) and considering 20% of balancing rate with Synthetic Minority Over-sampling Technique (SMOTE). Finally, our findings of most false negatives for laryngitis may explain the reason why its detection is a serious issue in speech technology. The results we report provide an original contribution, allowing for the consistent detection of multiple speech pathologies and advancing the state-of-the-art in the field of handcrafted acoustic-based non-invasive diagnosis of voice disorders.