Kanchi Mamunivar Centre for Post Graduate Studies

The study introduces cross-modal memory-enhanced fusion (CrossMF) as a unified transformer-based model which performs emotion recognition from speech and text data. CrossMF combines dynamic attention between modalities with a memory enhancement system for effective fusion between textual and acoustic information. The training of CrossMF involves a two-step tactic where the acoustic encoder receives clean audio inputs from the Toronto emotional speech set (TESS) for optimization, yet the text encoder and fusion module acquire training from audio-text pairs from the multimodal EmotionLines (MELD) dataset. The system allows emotion predictions from three combination types, including audio-only, text-only, and audio with text inputs, due to adjustable modality access. The evaluation process takes place across the three different input scenarios to show that the system performs well with high generalization capability in both laboratory-recorded and natural conversational conditions. Fusion between multiple sources only occurs when both inputs are available so that integrity can be maintained until integration occurs. The model reaches a maximum validation accuracy of 97.68% while demonstrating sustained high-test performance, which proves its effectiveness when operated in various conditions without depending on manually created features. The architecture also supports future extensions, allowing developers to easily incorporate ablation studies and adaptive training strategies for real-world emotion-aware systems.

Morphometric and meristic traits offer powerful, cost-effective tools for analysing phenotypic variation, stock structure, and environmental adaptation in freshwater fishes. This review synthesizes findings from more than thirty studies published between 2012 and 2024, covering diverse taxa across Asia, Africa, and the Middle East. The literature reveals strong habitat-driven morphological plasticity, particularly in riverine species exposed to hydrodynamic forces, ecological gradients, and climatic variability. Morphometric traits, which respond rapidly to ecological pressures, consistently differentiated populations of Channa striata, Wallago attu, Schizothorax spp., Clupisoma garua, and several others. Meristic traits, although more genetically stable, also reflected population-level divergence under varied developmental and environmental conditions. Integrating these patterns, the review highlights the importance of morphology for taxonomy, species discovery, biodiversity monitoring, and sustainable fisheries management. Despite methodological challenges—including inconsistent measurement protocols and limited geographic coverage—advances in geometric morphometrics, molecular barcoding, and multivariate analytics have strengthened morphological research. The review emphasizes the need for standardized methodologies and integrated approaches to better understand phenotypic diversity and ensure the conservation of freshwater fish resources in rapidly changing ecosystems.

In this paper, we propose a convolutional neural network (CNN) and Bidirectional Long Short-Term Memory (Bi-LSTM) based deep learning for speech emotion classification, which uses the MFCC to extract features. Data augmentation and the merging of popular datasets, RAVDESS (Ryerson Audio-Visual Database of Emotional Speech and Song) and TESS (Toronto Emotional Speech Set) used to enhance generalization, while Bayesian Optimization tunes hyperparameters. Experimental results of the proposed model achieved 99.77% training accuracy, 97.78% validation accuracy, and 98.34% testing accuracy, demonstrating the effectiveness of the model.

This work aims to produce semiconductor nanoparticles capable of harnessing visible light for the degradation of dyes and microbes. Employing an ionic liquid-assisted sol-gel process with varying dopant weight percentages, the study focuses on crafting Cerium (Ce) and Phosphorus (P) doped TiO2 Nanomaterials. Structural assessments, including Powder X-ray Diffraction (confirming the anatase phase), Transmission Electron Microscopy (revealing a particle size of 6.2 nm), Brunauer-Emmett-Teller surface area analysis (yielding 166 m2/gr), and Scanning Electron Microscopy (examining the morphology), were conducted. The catalysts were further evaluated for optical characteristics: UV-vis diffuse reflectance spectrum (indicating an energy gap of 2.59 eV), Electrochemical Impedance Spectroscopy (with an Efb of -0.30 V), and Valence band XPS (showing Evb at 2.03 eV). Substitutional doping of dopants into the TiO2 lattice was confirmed through X-ray photoelectron spectroscopy and Fourier Transform Infra-Red analysis. Photoluminescence Spectrum and Time Correlated Single Photon Counting analysis was employed to investigate electron-hole recombination. These characterizations suggest the catalysts are effective in degrading microorganisms and dyes under visible light exposure. Optimal conditions were obtained using CPT5IL2 at pH 3, 0.10 g catalyst dosage, and an initial dye concentration of 10 mg/L, which were determined to achieve complete dye degradation within 60 min. Furthermore, the catalyst's antibacterial and antifungal activity against Enterobacter aerogenes (MTCC-241, Gram-negative), Salmonella typhimurium (MTCC-98, Gram-negative), and Candida albicans (MTCC-277) were studied.

Phthalazine derivatives are an essential component of biological systems that serve as a structural template for biologically active molecules. In this chapter, Novel N-(4-(1,2-dihydrophthalazin-1-yl) phenyl) amides were synthesized in good yields, beginning with inexpensive and easily accessible phthalic anhydride using Pd (OAc)2 catalyzed Suzuki-mayura coupling and N-acylation reactions as key steps. All the prepared compounds were characterized by various spectroscopic techniques such as 1H NMR, 13C NMR, IR, and LCMS and all the data showed good agreement with the predicted structures. Additionally, the produced compounds' in vitro anti-microbial activity was examined. In comparison to Xanthomonas campestris, the target compounds 7i, 7f, and 7a demonstrated superior activity. In contrast, the compounds 7j, 7a, and 7h demonstrated excellent activity against Bacillus megaterium. Escherichia coli was effectively inhibited by compounds 7j and 7a. The obtained N-(4-(1,2-dihydrophthalazin-1-yl) phenyl) amide derivatives offer an intriguing profile in the field of antimicrobial research.

This study investigated the medicinal potential of Lavatera cashmeriana, a plant traditionally known for its therapeutic properties. The aim was to identify the phytocompounds in L. cashmeriana leaf extract and evaluate its antibacterial, antioxidant, and anticancer effects. Gas chromatography-mass spectrometry analysis was employed to characterize the phytochemical composition of the ethanol extract derived from L. cashmeriana leaves. The antimicrobial potential was assessed through the well diffusion technique, targeting Escherichia coli, Enterococcus faecalis, Pseudomonas aeruginosa, Staphylococcus aureus, and Candida albicans. The 2,2-diphenyl-1-picrylhydrazyl assay was conducted to assess antioxidant capabilities, while cytotoxicity against the A549 cancer cell line was determined via the MTT assay. GC-MS analysis identified ten different compounds, with phytol, 1-Eicosanol, and 2,6,10-trimethyl,14-ethylene-14-pentadecne being the most prevalent. The extract exhibited notable antimicrobial efficacy against all bacteria with MIC values ranging from 62.5 to 250 µg/mL. However, C. albicans did not respond. The extract exhibited antioxidative properties with an IC50 value of 86 µg/mL and cytotoxicity with an IC50 value of 69.95 µg/mL against the A549 cancer cell line. The results derived from this study supported the historical use of L. cashmeriana as a medicinal plant and suggested that it can potentially treat a wide range of medical ailments. The identified phytocompounds and the demonstrated antibacterial, antioxidant, and anticancer effects provide scientific evidence for its medicinal properties. However, further investigations are needed to fully understand its safety profile, efficacy, and mechanism of action before recommending it for therapeutic purposes.