Ahnaf Mozib Samin

Ahnaf Mozib Samin, M. Humayon Kobir, Md. Mushtaq Shahriyar Rafee et al.

Despite huge improvements in automatic speech recognition (ASR) employing neural networks, ASR systems still suffer from a lack of robustness and generalizability issues due to domain shifting. This is mainly because principal corpus design criteria are often not identified and examined adequately while compiling ASR datasets. In this study, we investigate the robustness of the state-of-the-art transfer learning approaches such as self-supervised wav2vec 2.0 and weakly supervised Whisper as well as fully supervised convolutional neural networks (CNNs) for multi-domain ASR. We also demonstrate the significance of domain selection while building a corpus by assessing these models on a novel multi-domain Bangladeshi Bangla ASR evaluation benchmark - BanSpeech, which contains approximately 6.52 hours of human-annotated speech and 8085 utterances from 13 distinct domains. SUBAK.KO, a mostly read speech corpus for the morphologically rich language Bangla, has been used to train the ASR systems. Experimental evaluation reveals that self-supervised cross-lingual pre-training is the best strategy compared to weak supervision and full supervision to tackle the multi-domain ASR task. Moreover, the ASR models trained on SUBAK.KO face difficulty recognizing speech from domains with mostly spontaneous speech. The BanSpeech will be publicly available to meet the need for a challenging evaluation benchmark for Bangla ASR.

Ahnaf Mozib Samin, Behrooz Nikandish, Jingyan Chen

Handling and digesting a huge amount of information in an efficient manner has been a long-term demand in modern society. Some solutions to map key points (short textual summaries capturing essential information and filtering redundancies) to a large number of arguments/opinions have been provided recently (Bar-Haim et al., 2020). To complement the full picture of the argument-to-keypoint mapping task, we mainly propose two approaches in this paper. The first approach is to incorporate prompt engineering for fine-tuning the pre-trained language models (PLMs). The second approach utilizes prompt-based learning in PLMs to generate intermediary texts, which are then combined with the original argument-keypoint pairs and fed as inputs to a classifier, thereby mapping them. Furthermore, we extend the experiments to cross/in-domain to conduct an in-depth analysis. In our evaluation, we find that i) using prompt engineering in a more direct way (Approach 1) can yield promising results and improve the performance; ii) Approach 2 performs considerably worse than Approach 1 due to the negation issue of the PLM.

Md Zobaer Hossain, Ahnaf Mozib Samin

The extent to which men and women use language differently has been questioned previously. Finding clear and consistent gender differences in language is not conclusive in general, and the research is heavily influenced by the context and method employed to identify the difference. In addition, the majority of the research was conducted in written form, and the sample was collected in writing. Therefore, we compared the word choices of male and female presenters in public addresses such as TED lectures. The frequency of numerous types of words, such as parts of speech (POS), linguistic, psychological, and cognitive terms were analyzed statistically to determine how male and female speakers use words differently. Based on our data, we determined that male speakers use specific types of linguistic, psychological, cognitive, and social words in considerably greater frequency than female speakers.

Ahnaf Mozib Samin, M. Firoz Ahmed, Md. Mushtaq Shahriyar Rafee

With the utilization of Transformer architecture, large Vision and Language (V&L) models have shown promising performance in even zero-shot settings. Several studies, however, indicate a lack of robustness of the models when dealing with complex linguistics and visual attributes. In this work, we introduce a novel V&L benchmark - ColorFoil, by creating color-related foils to assess the models' perception ability to detect colors like red, white, green, etc. We evaluate seven state-of-the-art V&L models including CLIP, ViLT, GroupViT, and BridgeTower, etc. in a zero-shot setting and present intriguing findings from the V&L models. The experimental evaluation indicates that ViLT and BridgeTower demonstrate much better color perception capabilities compared to CLIP and its variants and GroupViT. Moreover, CLIP-based models and GroupViT struggle to distinguish colors that are visually distinct to humans with normal color perception ability.

Aditya Mishra, Ahnaf Mozib Samin, Ali Etemad et al.

We propose GC-VASE, a graph convolutional-based variational autoencoder that leverages contrastive learning for subject representation learning from EEG data. Our method successfully learns robust subject-specific latent representations using the split-latent space architecture tailored for subject identification. To enhance the model's adaptability to unseen subjects without extensive retraining, we introduce an attention-based adapter network for fine-tuning, which reduces the computational cost of adapting the model to new subjects. Our method significantly outperforms other deep learning approaches, achieving state-of-the-art results with a subject balanced accuracy of 89.81% on the ERP-Core dataset and 70.85% on the SleepEDFx-20 dataset. After subject adaptive fine-tuning using adapters and attention layers, GC-VASE further improves the subject balanced accuracy to 90.31% on ERP-Core. Additionally, we perform a detailed ablation study to highlight the impact of the key components of our method.

Ahnaf Mozib Samin

Byte pair encoding (BPE) emerges as an effective tokenization method for tackling the out-of-vocabulary (OOV) challenge in various natural language and speech processing tasks. Recent research highlights the dependency of BPE subword tokenization's efficacy on the morphological nature of the language, particularly in languages rich in inflectional morphology, where fewer BPE merges suffice for generating highly productive tokens. Motivated by this, our study empirically identifies the optimal number of BPE tokens for Bengali, a language known for its morphological complexity, thus enhancing out-of-distribution automatic speech recognition (ASR) performance. Experimental evaluation reveals that an excessively high number of BPE tokens can lead to overfitting, while approximately 500-1000 tokens result in superior OOV performance. Furthermore, we conduct a comparative analysis of BPE with character-based and unigram-based tokenization methods. By introducing BPE tokenization to Bengali ASR, we achieve a substantial reduction in the word error rate (WER) from 66.44% in our character-based baseline system to 63.80% on the LB-ASRTD eval set and from 46.34% to 42.80% on the SHRUTI eval set, both of which include out-of-distribution data.