Explore the vast range of titles available.

This book offers a comprehensive account of the formal and semantic aspects of the two most prominent voice phenomena in Marathi, viz. the passive and the causative in the functional-typological framework.

Vocal biomarker-based machine learning approaches have shown promising results in the detection of various health conditions, including respiratory diseases, such as asthma. This study aimed to determine whether a respiratory-responsive vocal biomarker (RRVB) model platform initially trained on an asthma and healthy volunteer (HV) data set can differentiate patients with active COVID-19 infection from asymptomatic HVs by assessing its sensitivity, specificity, and odds ratio (OR). A logistic regression model using a weighted sum of voice acoustic features was previously trained and validated on a data set of approximately 1700 patients with a confirmed asthma diagnosis and a similar number of healthy controls. The same model has shown generalizability to patients with chronic obstructive pulmonary disease, interstitial lung disease, and cough. In this study, 497 participants (female: n=268, 53.9%; <65 years old: n=467, 94%; Marathi speakers: n=253, 50.9%; English speakers: n=223, 44.9%; Spanish speakers: n=25, 5%) were enrolled across 4 clinical sites in the United States and India and provided voice samples and symptom reports on their personal smartphones. The participants included patients who are symptomatic COVID-19 positive and negative as well as asymptomatic HVs. The RRVB model performance was assessed by comparing it with the clinical diagnosis of COVID-19 confirmed by reverse transcriptase-polymerase chain reaction. The ability of the RRVB model to differentiate patients with respiratory conditions from healthy controls was previously demonstrated on validation data in asthma, chronic obstructive pulmonary disease, interstitial lung disease, and cough, with ORs of 4.3, 9.1, 3.1, and 3.9, respectively. The same RRVB model in this study in COVID-19 performed with a sensitivity of 73.2%, specificity of 62.9%, and OR of 4.64 (P<.001). Patients who experienced respiratory symptoms were detected more frequently than those who did not experience respiratory symptoms and completely asymptomatic patients (sensitivity: 78.4% vs 67.4% vs 68%, respectively). The RRVB model has shown good generalizability across respiratory conditions, geographies, and languages. Results using data set of patients with COVID-19 demonstrate its meaningful potential to serve as a prescreening tool for identifying individuals at risk for COVID-19 infection in combination with temperature and symptom reports. Although not a COVID-19 test, these results suggest that the RRVB model can encourage targeted testing. Moreover, the generalizability of this model for detecting respiratory symptoms across different linguistic and geographic contexts suggests a potential path for the development and validation of voice-based tools for broader disease surveillance and monitoring applications in the future.

India’s linguistic diversity encompasses multiple language families, including the Indo-Aryan and Dravidian, which represent distinct phonological and morphological characteristics. This study aims to evaluate and compare the performance of end-to-end automatic speech recognition (ASR) systems for three Indo-Aryan languages—Marathi, Odia, and Gujarati—and three Dravidian languages—Tamil, Telugu, and Malayalam. Using four transformer-based pre-trained models—Wav2Vec2.0-base, XLSR-53, W2V2-BERT, and Whisper small—the analysis explores their adaptability to these languages’ linguistic features, with word error rate (WER) and character error rate (CER) serving as evaluation metrics. Results indicate that W2V2-BERT and XLSR-53 outperform other models, achieving lower WER and CER, especially for Indo-Aryan languages. However, higher error rates for Dravidian languages highlight challenges such as complex phonology and agglutinative morphology. This work provides a comparative insight into the strengths and limitations of pre-trained ASR models across India’s diverse linguistic landscape and underscores the need for language-specific adaptations to improve ASR accuracy for underrepresented languages.

Decompounding is an essential preprocessing step in text-processing tasks such as machine translation, speech recognition, and information retrieval (IR). Here, the IR issues are explored from five viewpoints. (A) Does word decompounding impact the Indian language IR? If yes, to what extent? (B) Can corpus-based decompounding models be used in the Indian language IR? If yes, how? (C) Can machine learning and deep learning-based decompounding models be applied in the Indian language IR? If yes, how? (D) Among the different decompounding models (corpus-based, hybrid machine learning-based, and deep learning-based), which provides the best effectiveness in the IR domain? (E) Among the different IR models, which provides the best effectiveness from the IR perspective? This study proposes different corpus-based, hybrid machine learning-based, and deep learning-based decompounding models in Indian languages (Marathi, Hindi, and Sanskrit). Moreover, we evaluate the effectiveness of each activity from an IR perspective only. It is observed that the different decompounding models improve IR effectiveness. The deep learning-based decompounding models outperform the corpus-based and hybrid machine learning-based models in Indian language IR. Among the different deep learning-based models, the Bi-LSTM-A model performs best and improves mean average precision (MAP) by 28.02% in Marathi. Similarly, the Bi-RNN-A model improves MAP by 18.18% and 6.1% in Hindi and Sanskrit, respectively. Among the retrieval models, the In_expC2 model outperforms others in Marathi and Hindi, and the BB2 model outperforms others in Sanskrit.

Stress is defined as a form of psychalgia. Owing to the current day lifestyle of Homo-sapiens, the most recurring pain is psychogenic; and the most damaging form of psychalgia. Stress in its most severe form, has led to the death of many individuals of this species. In accordance to a study conducted by WHO in 2015, around 800,000 individuals commit suicide each year (one individual per 40 s). The only solution to this conundrum is to bring in efficient mechanized stress detection technique which utilize proven measures and are unbiased, is called “speech emotion recognition” (SER). Stress, by itself, is not an emotion, but gives rise to specific emotions. This paper proposes SER using neural network classifier with weight optimization using fusion of optimization algorithms viz. BAT, genetic algorithm, particle swarm organization and simulated annealing. Classifier is trained using multi-model feature set. Gammatone Wavelet Cepstral coefficient, Mel Frequency Cepstral coefficient, pitch, vocal tract frequency and energy are the features used to identify different emotions. Detect the stress level being main objective SUSAS benchmark database and Marathi language database is used for performance analysis. Performance parameters like cost function for evaluating meta-heuristic optimization algorithm and accuracy of emotion detection is calculated. The overall accuracy of 84.2% of stress related emotions is achieved.

Hidden Markov Model and Deep Neural Networks based Statistical Parametric Speech Synthesis systems, gain a significant attention from researchers because of their flexibility in generating speech waveforms in diverse voice qualities as well as in styles. This paper describes HMM-based speech synthesis system (SPSS) for the Marathi language. In proposed synthesis method, speech parameter trajectories used for synthesis are generated from the trained hidden Markov models (HMM). We have recorded our database of 5300 phonetically balanced Marathi sentences to train the context-dependent HMM with five, seven and nine hidden states. The subjective quality measures (MOS and PWP) shows that the HMMs with seven hidden states are capable of giving an adequate quality of synthesized speech as compared to five state and with less time complexity than seven state HMMs. The contextual features used for experimentation are inclusive of a position of an observed phoneme in a respective syllable, word, and sentence.

Automatic Speaker Recognition (ASR) is an economic tool for voice biometrics because of availability of low cost and powerful processors. For an ASR system to be successful in practical environments, it must have high mimic resistance , i.e., the system should not be defeated by determined mimics which may be either identical twins or professional mimics. In this paper, we demonstrate the effectiveness of Linear Prediction (LP)-based features, viz., Linear Prediction Coefficients (LPC) and Linear Prediction Cepstral Coefficients (LPCC) over filterbank-based features such as Mel-Frequency Cepstral Coefficients (MFCC) and newly proposed Teager energy-based MFCC (T-MFCC) for the identification of professional mimics in Indian languages. Results are reported for real and fictitious experiments. On the whole, it is observed that LP-based features perform better than filterbank-based features (an average jump of 23.21% and 31.43% for fictitious experiments with professional mimic in Marathi and Hindi, respectively, whereas there is an average jump of 1.64% for real experiments with professional mimic in Hindi) and we believe that this is the first time such results on identification of professional mimics in ASR are obtained . Analysis of the results is given with the help of Mean Square Error (MSE) between training and testing utterances for mimic’s imitations for target speakers and target speakers’ normal voice. Fourier spectra and corresponding LP spectra for target speaker and its impersonations provided by professional mimic are shown to justify the results. Finally, dependence of LPC on physiological characteristics of vocal tract and its relation with respect to the problem addressed in this paper is studied.

Automatic Speaker Recognition (ASR) refers to the task of identifying a person based on his or her voice with the help of machines. ASR finds its potential applications in telephone based financial transactions, purchase of credit card and in forensic science and social anthropology for the study of different cultures and languages. Results of ASR are highly dependent on database, i.e., the results obtained in ASR are meaningless if recording conditions are not known. In this paper, a methodology and a typical experimental setup used for development of corpora for various tasks in the text-independent speaker identification in different Indian languages, viz., Marathi, Hindi, Urdu and Oriya have been described. Finally, an ASR system is presented to evaluate the corpora.