Amalesh Gope
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Amalesh Gope has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Anusuya Pal
The University of Tokyo
5 shared publications Dinkur Borah
Tezpur University
2 shared publications- Germano S. Iannacchione
Order-Disorder Phenomena Laboratory, Department of Physics, Worcester Polytechnic Institute, Worcester, MA 01609, USA
2 shared publications Sekholu Tetseo
Department of Linguistics and Language Technology, Tezpur University, Tezpur 784028, Assam, India
2 shared publications- Shakuntala Mahanta
IIT Guwahati
2 shared publications Tulika Gogoi
Department of Linguistics and Language Technology, Tezpur University, Tezpur 784028, Assam, Indiaa
2 shared publications
Latest Publications
- Open Access
Morphological signatures of bacteria‐laden drying droplets: A multimodal analysis across host fluids
Drying droplets exhibit rich morphological patterns that depend on fluid composition, yet their use in characterizing bacteria‐laden fluids remains underexplored. This paper introduces a multimodal machine learning (ML) framework that integrates numerical descriptors of texture evolution with image‐based features to examine how drying patterns vary with the concentration of Escherichia coli bacteria in three host fluids, viz., blood serum, urine, and nutrient‐rich medium. As a proof‐of‐concept, our approach exploits drying‐induced self‐assembly in bacteria‐laden host fluids as an analytical signal, providing a label‐free route to estimate bacterial concentration from microliter‐scale droplets under controlled laboratory conditions. Time‐resolved fluorescence imaging of green fluorescent protein (GFP)‐expressing bacteria further reveals distinct spatiotemporal redistribution dynamics during droplet evaporation across different fluids, providing a physical basis for the unique morphological signatures captured in the dried patterns. Rather than relying on large deep‐learning backbones, we demonstrate that optimal classification accuracy arises from a controlled balance between semantic and textural information within the image encoder. Systematic depth‐truncation experiments reveal an “optimal representational zone” where texture‐rich and biologically meaningful features combine with numerical descriptors, providing accuracy under frame‐level evaluation, with accuracy retained when tested on entirely held‐out droplets, indicating that the model captures reproducible morphological features beyond frame‐level temporal correlation. Gradient‐weighted class activation mapping (Grad‐CAM) confirms that the model attends to physically interpretable morphological cues linked to bacterial concentrations, consistent with drying droplet morphology encoding reproducible signatures across three compositionally distinct host fluids.
Open AccessMorphological signatures of bacteria‐laden drying droplets: A multimodal analysis across host fluids
Drying droplets exhibit rich morphological patterns that depend on fluid composition, yet their use in characterizing bacteria‐laden fluids remains underexplored. This paper introduces a multimodal machine learning (ML) framework that integrates numerical descriptors of texture evolution with image‐based features to examine how drying patterns vary with the concentration of Escherichia coli bacteria in three host fluids, viz., blood serum, urine, and nutrient‐rich medium. As a proof‐of‐concept, our approach exploits drying‐induced self‐assembly in bacteria‐laden host fluids as an analytical signal, providing a label‐free route to estimate bacterial concentration from microliter‐scale droplets under controlled laboratory conditions. Time‐resolved fluorescence imaging of green fluorescent protein (GFP)‐expressing bacteria further reveals distinct spatiotemporal redistribution dynamics during droplet evaporation across different fluids, providing a physical basis for the unique morphological signatures captured in the dried patterns. Rather than relying on large deep‐learning backbones, we demonstrate that optimal classification accuracy arises from a controlled balance between semantic and textural information within the image encoder. Systematic depth‐truncation experiments reveal an “optimal representational zone” where texture‐rich and biologically meaningful features combine with numerical descriptors, providing accuracy under frame‐level evaluation, with accuracy retained when tested on entirely held‐out droplets, indicating that the model captures reproducible morphological features beyond frame‐level temporal correlation. Gradient‐weighted class activation mapping (Grad‐CAM) confirms that the model attends to physically interpretable morphological cues linked to bacterial concentrations, consistent with drying droplet morphology encoding reproducible signatures across three compositionally distinct host fluids.
- Open Access
- 9 April 2026
- Published by John Benjamins Publishing Company
- Vol. 27 (4) , 645-684
Exploring tonal contrasts in Dharamshala Tibetan using mathematical modeling, statistical modeling, and machine learning algorithms
This paper examines the tonal system of Dharamshala Tibetan (DT) through a comprehensive phonetic and phonological analysis, employing statistical tests, regression modeling, mathematical modeling, and machine learning algorithms (MLAs). Our aim has been to determine the number and nature of tonal contrasts in DT, an issue that remains unresolved to date. The stimuli (tonemes), consisting of two-way and three-way meaning contrasts, are designed based on preliminary fieldwork and consultations with native speakers. Acoustic analyses reveal three distinct tonal categories supported by significant differences in fundamental frequency (f0), including f0 height and f0 directionality, duration, and intensity. Although initial f0 data analysis, particularly f0 directionality, suggested rising contours, further analysis confirmed these as (i) two level tones, viz . high (H) and low (L), and (ii) one contour tone, viz. , mid-rising (MR), with consistent pitch trajectories. MLAs were employed to classify the tones, with confusion matrices highlighting the prominence of misclassifications involving mid-rising tones. This supports the hypothesis that the mid-rising tone may be a recently emerged category. The findings of this work are contextualized within the broader framework of Tibetan phonology, providing valuable insights into the tonal characteristics of DT. This study underscores the importance of integrating quantitative approaches with linguistic theory to deepen our understanding of tonal systems.
- 9 April 2026
- Published by John Benjamins Publishing Company
- Vol. 27 (4) , 645-684
Open AccessExploring tonal contrasts in Dharamshala Tibetan using mathematical modeling, statistical modeling, and machine learning algorithms
This paper examines the tonal system of Dharamshala Tibetan (DT) through a comprehensive phonetic and phonological analysis, employing statistical tests, regression modeling, mathematical modeling, and machine learning algorithms (MLAs). Our aim has been to determine the number and nature of tonal contrasts in DT, an issue that remains unresolved to date. The stimuli (tonemes), consisting of two-way and three-way meaning contrasts, are designed based on preliminary fieldwork and consultations with native speakers. Acoustic analyses reveal three distinct tonal categories supported by significant differences in fundamental frequency (f0), including f0 height and f0 directionality, duration, and intensity. Although initial f0 data analysis, particularly f0 directionality, suggested rising contours, further analysis confirmed these as (i) two level tones, viz . high (H) and low (L), and (ii) one contour tone, viz. , mid-rising (MR), with consistent pitch trajectories. MLAs were employed to classify the tones, with confusion matrices highlighting the prominence of misclassifications involving mid-rising tones. This supports the hypothesis that the mid-rising tone may be a recently emerged category. The findings of this work are contextualized within the broader framework of Tibetan phonology, providing valuable insights into the tonal characteristics of DT. This study underscores the importance of integrating quantitative approaches with linguistic theory to deepen our understanding of tonal systems.
- Open Access
- 7 January 2026
- Published by Springer Nature in International Journal of Speech Technology
- Vol. 29 (1) , 17
Speech-driven sentence type classification in Chokri using traditional and transfer learning methods
Sentence classification typically relies on textual information in high-resource languages. However, it is rather challenging to perform such tasks in under-resourced languages that often lack sufficient text or speech data to support such tasks. This paper examines the effectiveness of audio features derived from speech data (speech features) in the endangered tonal language Chokri for classifying sentence types using traditional machine learning and transfer learning algorithms. Results show that speech features can be effectively used for sentence classification tasks without relying on textual information. Two approaches are employed to achieve the sentence classification tasks. First, we extracted various individual and combined speech features, including Mel Frequency Cepstral Coefficients (MFCCs), Chroma, and other spectral features, as numerical data and applied the random forest algorithm for traditional machine learning. The results show that the average F1 score for MFCCs alone, MFCCs with other features, and MFCCs with Chroma is 86% each, with the highest average of 88% achieved by combining all features. These findings underscore the critical role of MFCCs in speech-based sentence classification and the varying contributions of other features. The second approach involves converting speech into spectrogram and MFCC images and applying transfer learning models like ResNet101 and InceptionResNet. This method improves classification accuracy by 10%, resulting in 97–98% accuracy. This study suggests further experimentation with various grammatical structures by demonstrating the effectiveness of speech features alone for sentence classification, thereby expanding the possibilities for language processing tasks in resource-limited scenarios.
- 7 January 2026
- Published by Springer Nature in International Journal of Speech Technology
- Vol. 29 (1) , 17
Open AccessSpeech-driven sentence type classification in Chokri using traditional and transfer learning methods
Sentence classification typically relies on textual information in high-resource languages. However, it is rather challenging to perform such tasks in under-resourced languages that often lack sufficient text or speech data to support such tasks. This paper examines the effectiveness of audio features derived from speech data (speech features) in the endangered tonal language Chokri for classifying sentence types using traditional machine learning and transfer learning algorithms. Results show that speech features can be effectively used for sentence classification tasks without relying on textual information. Two approaches are employed to achieve the sentence classification tasks. First, we extracted various individual and combined speech features, including Mel Frequency Cepstral Coefficients (MFCCs), Chroma, and other spectral features, as numerical data and applied the random forest algorithm for traditional machine learning. The results show that the average F1 score for MFCCs alone, MFCCs with other features, and MFCCs with Chroma is 86% each, with the highest average of 88% achieved by combining all features. These findings underscore the critical role of MFCCs in speech-based sentence classification and the varying contributions of other features. The second approach involves converting speech into spectrogram and MFCC images and applying transfer learning models like ResNet101 and InceptionResNet. This method improves classification accuracy by 10%, resulting in 97–98% accuracy. This study suggests further experimentation with various grammatical structures by demonstrating the effectiveness of speech features alone for sentence classification, thereby expanding the possibilities for language processing tasks in resource-limited scenarios.
- 3 June 2025
- Published by Springer Nature
- p. 55-71
The Constraints of Linguistic Fieldwork and Documentation Concerning the Languages of Nagaland
- 3 June 2025
- Published by Springer Nature
- p. 55-71
The Constraints of Linguistic Fieldwork and Documentation Concerning the Languages of Nagaland
- Open Access
- 14 November 2024
- Published by Bilingual Publishing Group in Forum for Linguistic Studies
- Vol. 6 (5) , 664-677
A Mathematical and Statistical Approach to Explore the Downtrend Properties in Chokri
This paper explores f0 downtrend processes in Chokri, a Tibeto-Burman language spoken in Nagaland, India. Chokri manifests five lexical tones, comprising four level tones- extra high (EH), high (H), mid (M), low (L), and one contour tone- mid-rising (MR). This study encompasses three distinct downtrend phenomena: temporal declination, final lowering, and automatic downstep. Three complementary methodologies- (a) visual, (b) statistical, and (c) mathematical modeling, are incorporated to examine time-dependent declination across all H, M, and L tone sequences. Visual and statistical analyses show the downtrend nature of these tonal sequences, revealing a linear downtrend in all H and L tone sequences, with M tone displaying a non-linear trajectory. However, while visual observation and statistical modeling do not precisely delineate the appropriate non-linear model, mathematical modeling confirms exponential decay as the fitting non-linear model for M tone sequences. The mathematical modeling further validates that only M tone sequences result in final lowering. Furthermore, an intervening L tone affects the f0 scaling of H tone sequences, confirming an automatic downstep in Chokri. Quantitative analysis revealed the pronounced effect of the intervening L tone on the preceding and following syllables containing an H tone and confirmed the declination process in Chokri.
- 14 November 2024
- Published by Bilingual Publishing Group in Forum for Linguistic Studies
- Vol. 6 (5) , 664-677
Open AccessA Mathematical and Statistical Approach to Explore the Downtrend Properties in Chokri
This paper explores f0 downtrend processes in Chokri, a Tibeto-Burman language spoken in Nagaland, India. Chokri manifests five lexical tones, comprising four level tones- extra high (EH), high (H), mid (M), low (L), and one contour tone- mid-rising (MR). This study encompasses three distinct downtrend phenomena: temporal declination, final lowering, and automatic downstep. Three complementary methodologies- (a) visual, (b) statistical, and (c) mathematical modeling, are incorporated to examine time-dependent declination across all H, M, and L tone sequences. Visual and statistical analyses show the downtrend nature of these tonal sequences, revealing a linear downtrend in all H and L tone sequences, with M tone displaying a non-linear trajectory. However, while visual observation and statistical modeling do not precisely delineate the appropriate non-linear model, mathematical modeling confirms exponential decay as the fitting non-linear model for M tone sequences. The mathematical modeling further validates that only M tone sequences result in final lowering. Furthermore, an intervening L tone affects the f0 scaling of H tone sequences, confirming an automatic downstep in Chokri. Quantitative analysis revealed the pronounced effect of the intervening L tone on the preceding and following syllables containing an H tone and confirmed the declination process in Chokri.
- Open Access
- 10 May 2024
- Published by Springer Nature in Humanities and Social Sciences Communications
- Vol. 11 (1) , 1-13
Multi-class identification of tonal contrasts in Chokri using supervised machine learning algorithms
This study examines and explores the effectiveness of various Machine Learning Algorithms (MLAs) in identifying intricate tonal contrasts in Chokri (ISO 639-3), an under-documented and endangered Tibeto-Burman language of the Sino-Tibetan language family spoken in Nagaland, India. Seven different supervised MLAs, viz., [Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Naive Bayes (NB)], and one neural network (NN)-based algorithms [Artificial Neural Network (ANN)] are implemented to explore five-way tonal contrasts in Chokri. Acoustic correlates of tonal contrasts, encompassing fundamental frequency fluctuations, viz., f0 height and f0 direction, are examined. Contrary to the prevailing notion of NN supremacy, this study underscores the impressive accuracy achieved by the RF. Additionally, it reveals that combining f0 height and directionality enhances tonal contrast recognition for female speakers, while f0 directionality alone suffices for male speakers. The findings demonstrate MLAs’ potential to attain accuracy rates of 84–87% for females and 95–97% for males, showcasing their applicability in deciphering the intricate tonal systems of Chokri. The proposed methodology can be extended to predict multi-class problems in diverse fields such as image processing, speech classification, medical diagnosis, computer vision, and social network analysis.
- 10 May 2024
- Published by Springer Nature in Humanities and Social Sciences Communications
- Vol. 11 (1) , 1-13
Open AccessMulti-class identification of tonal contrasts in Chokri using supervised machine learning algorithms
This study examines and explores the effectiveness of various Machine Learning Algorithms (MLAs) in identifying intricate tonal contrasts in Chokri (ISO 639-3), an under-documented and endangered Tibeto-Burman language of the Sino-Tibetan language family spoken in Nagaland, India. Seven different supervised MLAs, viz., [Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Naive Bayes (NB)], and one neural network (NN)-based algorithms [Artificial Neural Network (ANN)] are implemented to explore five-way tonal contrasts in Chokri. Acoustic correlates of tonal contrasts, encompassing fundamental frequency fluctuations, viz., f0 height and f0 direction, are examined. Contrary to the prevailing notion of NN supremacy, this study underscores the impressive accuracy achieved by the RF. Additionally, it reveals that combining f0 height and directionality enhances tonal contrast recognition for female speakers, while f0 directionality alone suffices for male speakers. The findings demonstrate MLAs’ potential to attain accuracy rates of 84–87% for females and 95–97% for males, showcasing their applicability in deciphering the intricate tonal systems of Chokri. The proposed methodology can be extended to predict multi-class problems in diverse fields such as image processing, speech classification, medical diagnosis, computer vision, and social network analysis.
- Open Access
Hierarchical Exploration of Drying Patterns Formed in Drops Containing Lysozyme, PBS, and Liquid Crystals
Biological systems, by nature, are highly complex. These systems exhibit diverse hierarchical spatial and temporal features when driven far from equilibrium. The generated features are susceptible to the initial conditions that largely depend on vast parameter space. Extracting information on their properties and behavior thus becomes far too complex. This work seeks to examine the drying kinetics of the drops containing a globular protein (lysozyme (Lys)), phosphate buffer saline (PBS), and thermotropic liquid crystal (LCs). The drying evolution and the morphological crack patterns of these drops are examined using high-resolution microscopy, textural image analysis, and statistical methods. This study observes that the textural parameters can identify the (i) phase separation of the salts present in the PBS and (ii) the LCs’ birefringence during the drying evolution. This birefringence activities of the LCs slow down when the initial PBS concentration is increased from 0.25 to 1× despite using a fixed volume of LCs. To comprehend such a surprising effect, the combinations of (i) Lys+PBS and (ii) PBS+LCs are thoroughly examined. A phase diagram is established as a function of initial concentrations of Lys and PBS. The scanning electron microscopic images of Lys+PBS reveal that the tuning between lysozyme and salt concentrations in PBS plays a significant role in determining the morphological patterns. The Lys drops with and without LCs exhibit two distinct regions: the peripheral ring (“coffee-ring”) and the central ones. This phase-separated ring formation indicates that the film containing Lys and salts might have formed on top of these LCs in the central region, which reduces the optical response (birefringence) of LCs. A physical mechanism is proposed in this paper to anticipate the redistributions of LCs in a multi-component system such as Lys+PBS+LCs.
Open AccessHierarchical Exploration of Drying Patterns Formed in Drops Containing Lysozyme, PBS, and Liquid Crystals
Biological systems, by nature, are highly complex. These systems exhibit diverse hierarchical spatial and temporal features when driven far from equilibrium. The generated features are susceptible to the initial conditions that largely depend on vast parameter space. Extracting information on their properties and behavior thus becomes far too complex. This work seeks to examine the drying kinetics of the drops containing a globular protein (lysozyme (Lys)), phosphate buffer saline (PBS), and thermotropic liquid crystal (LCs). The drying evolution and the morphological crack patterns of these drops are examined using high-resolution microscopy, textural image analysis, and statistical methods. This study observes that the textural parameters can identify the (i) phase separation of the salts present in the PBS and (ii) the LCs’ birefringence during the drying evolution. This birefringence activities of the LCs slow down when the initial PBS concentration is increased from 0.25 to 1× despite using a fixed volume of LCs. To comprehend such a surprising effect, the combinations of (i) Lys+PBS and (ii) PBS+LCs are thoroughly examined. A phase diagram is established as a function of initial concentrations of Lys and PBS. The scanning electron microscopic images of Lys+PBS reveal that the tuning between lysozyme and salt concentrations in PBS plays a significant role in determining the morphological patterns. The Lys drops with and without LCs exhibit two distinct regions: the peripheral ring (“coffee-ring”) and the central ones. This phase-separated ring formation indicates that the film containing Lys and salts might have formed on top of these LCs in the central region, which reduces the optical response (birefringence) of LCs. A physical mechanism is proposed in this paper to anticipate the redistributions of LCs in a multi-component system such as Lys+PBS+LCs.
- Open Access
Pitch Range and Voice Quality in Dimasa Focus Intonation
This paper presents an analysis of Dimasa focus intonation. The acoustic analysis shows that narrow focus sentences undergo a jump in the pitch range irrespective of the underlying tonal value of the morpheme it attaches to. In addition to f0 expansion, the prosodic property of focus in Dimasa was found to have different (tense) phonation in morphologically marked narrow focus sentences when compared to the broad focus context. Thus, the tense phonation property of sentences bearing morphological focus is not only an acoustic property of a higher pitch range but may also be an acoustic cue of discourse-level intonation.
Open AccessPitch Range and Voice Quality in Dimasa Focus Intonation
This paper presents an analysis of Dimasa focus intonation. The acoustic analysis shows that narrow focus sentences undergo a jump in the pitch range irrespective of the underlying tonal value of the morpheme it attaches to. In addition to f0 expansion, the prosodic property of focus in Dimasa was found to have different (tense) phonation in morphologically marked narrow focus sentences when compared to the broad focus context. Thus, the tense phonation property of sentences bearing morphological focus is not only an acoustic property of a higher pitch range but may also be an acoustic cue of discourse-level intonation.
- Open Access
The Phonetics of Tone and Voice Quality Interactions in Sylheti
This paper examines the phonetic interactions of tone and voice qualities in Sylheti. Data from six native speakers are examined to understand the voice qualities of the vowels carrying contrastive tones. The results identify three spectral measures (viz., H1*–H2*, H1*–A2*, and H1*–A3*) and one noise measure (viz., CPP) as reliable indicators of modal (or in the continuum of modal to tense) vs. breathy (or, in the continuum of breathy to lax) phonation contrasts in the vowels carrying high and low tone, respectively. Finally, a statistical model is proposed that predicts consistent phonation contrasts across the total duration of the contrastive tones.
Open AccessThe Phonetics of Tone and Voice Quality Interactions in Sylheti
This paper examines the phonetic interactions of tone and voice qualities in Sylheti. Data from six native speakers are examined to understand the voice qualities of the vowels carrying contrastive tones. The results identify three spectral measures (viz., H1*–H2*, H1*–A2*, and H1*–A3*) and one noise measure (viz., CPP) as reliable indicators of modal (or in the continuum of modal to tense) vs. breathy (or, in the continuum of breathy to lax) phonation contrasts in the vowels carrying high and low tone, respectively. Finally, a statistical model is proposed that predicts consistent phonation contrasts across the total duration of the contrastive tones.
- Open Access
- 5 February 2021
- Published by MDPI AG in Biomolecules
- Vol. 11 (2) , 231
Temperature and Concentration Dependence of Human Whole Blood and Protein Drying Droplets
The drying of bio-colloidal droplets can be used in many medical and forensic applications. The whole human blood is the most complex bio-colloid system, whereas bovine serum albumin (BSA) is the simplest. This paper focuses on the drying characteristics and the final morphology of these two bio-colloids. The experiments were conducted by varying their initial concentrations, and the solutions were dried under various controlled substrate temperatures using optical and scanning electron microscopy. The droplet parameters (the contact angle, the fluid front, and the first-order image statistics) reveal the drying process’s unique features. Interestingly, both BSA and blood drying droplets’ contact angle measurements show evidence of a concentration-driven transition as the behavior changes from non-monotonic to monotonic decrease. This result indicates that this transition behavior is not limited to multi-component bio-colloid (blood) only, but may be a phenomenon of a bio-colloidal solution containing a large number of interacting components. The high dilution of blood behaves like the BSA solution. The ring-like deposition, the crack morphology, and the microstructures suggest that the components have enough time to segregate and deposit onto the substrate under ambient conditions. However, there is insufficient time for evaporative-driven segregation to occur at elevated temperatures, as expected.
- 5 February 2021
- Published by MDPI AG in Biomolecules
- Vol. 11 (2) , 231
Open AccessTemperature and Concentration Dependence of Human Whole Blood and Protein Drying Droplets
The drying of bio-colloidal droplets can be used in many medical and forensic applications. The whole human blood is the most complex bio-colloid system, whereas bovine serum albumin (BSA) is the simplest. This paper focuses on the drying characteristics and the final morphology of these two bio-colloids. The experiments were conducted by varying their initial concentrations, and the solutions were dried under various controlled substrate temperatures using optical and scanning electron microscopy. The droplet parameters (the contact angle, the fluid front, and the first-order image statistics) reveal the drying process’s unique features. Interestingly, both BSA and blood drying droplets’ contact angle measurements show evidence of a concentration-driven transition as the behavior changes from non-monotonic to monotonic decrease. This result indicates that this transition behavior is not limited to multi-component bio-colloid (blood) only, but may be a phenomenon of a bio-colloidal solution containing a large number of interacting components. The high dilution of blood behaves like the BSA solution. The ring-like deposition, the crack morphology, and the microstructures suggest that the components have enough time to segregate and deposit onto the substrate under ambient conditions. However, there is insufficient time for evaporative-driven segregation to occur at elevated temperatures, as expected.