This study introduces VACSR v.4.0, an updated version of the Vocabulary Analyzer for Self-Reflection. The tool automatically analyzes vocabulary occurrences and CEFR-J levels in transcribed texts, providing information on usage frequency, unused items, and tokens not classified within CEFR-J scales. VACSR v.4.0 incorporates syntactic parsing using the Stanza library, enabling the system to distinguish parts of speech and categorize identical lexical items according to their grammatical functions. The updated version also adds the automatic display of top headwords, a feature not included in previous versions, and extends its processing capability to tokens containing diacritical marks, such as ê and û, found in other languages. The study aims to explore how VACSR v.4.0 functions with languages other than English through a pilot analysis using two French texts containing vocabulary with diacritical marks. Although the system does not yet provide linguistically accurate POS tagging or CEFR-based level assignments for French, it can still extract high-frequency vocabulary and generate meaningful frequency-based lists. These findings show that, even without full multilingual NLP integration, VACSR v.4.0 helps compile core vocabulary lists across different languages and offers preliminary insights for pedagogical decision-making in foreign-language learning contexts.

The purpose of this paper is to record some unique and interesting linguistic features found in the Modern Bengali Cuisine Corpus (MBCC) and investigate how these features are represented in the cuisine texts of the corpus. The corpus is designed with curated cuisine texts obtained from various digital sources covering a small time window (2001-2024) with limited geographical variations. It contains only texts to reflect on some lexicosyntactic uniqueness of the text type. Since this is primarily a text corpus, the nontextual elements (e.g., images, graphs, sketches, videos and emojis) are removed from it to maximize textual representation and minimize linguistic inaccuracies. Subsequently, the corpus is put to some qualitative analyses to unveil some unique linguistic phenomena that are characteristically inherent to the culinary discourse. The initial analysis of the MBCC reveals the presence of cryptic syntactic structures, the prevalent use of imperative constructions, the high use of content words, frequent use of code-mixing and the strategies for targeted audience orientation. All these features, when considered together, underscore some inherent variabilities in linguistic structures dictated by communicative needs and manifest the remarkable proficiency of the human brain in extracting pertinent information from structurally complex texts. Further studies are necessary to understand the frequency of use of content words, distribution patterns of code-switching in discourse-diverse texts, linguistic strategies used in information encoding and the possibility of using this corpus in various domains of Bengali linguistics and technology developments.