ORIENTAL UNIVERSITETIQidiruvПоискSearch
Oriental universiteti nashriyotiIlmiy jurnallar platformasiПлатформа научных журналовScholarly journals platform

Toʻliq matnПолный текстFull text

Tilmoch.ai: bridging linguistic gaps with smart translation

Askarova Sabina

Matn PDF dan avtomatik ajratilgan: ekran oʻqiydigan dasturlar, telefon va qidiruv uchun. Rasmiy nusxa — PDF.Текст извлечён из PDF автоматически — для программ чтения с экрана, телефонов и поиска. Официальная версия — PDF.Text was extracted from the PDF automatically for screen readers, phones and search. The PDF remains the version of record.

PDF ni oʻqishЧитать PDFRead the PDF → Nashr kartasiКарточка изданияRecord

TILMOCH.AI: BRIDGING LINGUISTIC GAPS WITH SMART
TRANSLATION

https://zenodo.org/records/17732002

Askarova Sabina,
Profi Universiteti ingliz tili oʻqituvchisi

Abstract. This article analyzes Tilmoch.ai, an Uzbek-founded artificial
intelligence platform focusing in the translation of Central Asian and Turkic languages
such as Uzbek, Kazakh, and Turkish. Tilmoch.ai focuces on linguistic depth over
global platforms. The study looks at the benefits, drawbacks, and advantages of
translation education for local community empowerment, with a focus on translation
education especially. Although Tilmoch.ai provides translations that are both
technically correct and sensitive, the results demonstrate that human involvement is
still crucial, particularly in literary and idiomatic contexts. It comes to the conclusion
that translation pedagogy skills are still crucial, which means that Tilmoch AI can assist
translators in minimizing down on their translation time if it is implemented
intelligently and sensibly.

Keywords: artificial intelligence, translation, Tilmoch.ai, Turkic languages,
digital inclusion

Annotatsiya. Ushbu maqolada oʻzbek, qozoq va turk kabi Markaziy Osiyo va
turkiy tillarni tarjima qilishga ixtisoslashgan Tilmoch.ai sunʼiy intellekt platformasi
tahlil qilinadi. Tilmoch.ai global platformalarda lingvistik chuqurlikka eʼtibor qaratadi.
Tadqiqotda mahalliy hamjamiyat imkoniyatlarini kengaytirish uchun tarjima
taʼlimining afzalliklari, kamchiliklari va afzalliklari koʻrib chiqiladi, ayniqsa tarjima
sohasiga eʼtibor qaratiladi. Tilmoch.ai ham texnik jihatdan toʻgʻri, ham nozik
tarjimalarni taqdim etsa-da, natijalar shuni koʻrsatadiki, inson ishtiroki, ayniqsa adabiy
va idiomatik kontekstlarda, hali ham muhim ahamiyatga ega.Bundan shunday xulosa
kelib chiqadiki, tarjima pedagogikasi koʻnikmalari hali ham muhim ahamiyatga ega,
yaʼni Tilmoch AI tarjimonlarga, agar u aqlli va oqilona amalga oshirilsa, tarjima
vaqtini minimallashtirishda yordam berishi mumkin.

Kalit soʻzlar: sunʼiy intellekt, tarjima, Tilmoch.ai, turkiy tillar, raqamli
inklyuziya.

Introduction. In todayʼs interconnected world, language barriers continue to
impede equitable access to knowledge and opportunity. Global translation systems
such as Google Translate and DeepL offer wide linguistic coverage but often perform
poorly with underrepresented languages. This digital exclusion affects not only
communication but also the preservation of cultural identity.

Tilmoch.ai, launched in 2023 in Uzbekistan, seeks to address this gap by
prioritizing regional languages of Central Asia. This article evaluates its contribution
to translation technology, explores its practical applications, and identifies limitations
that necessitate continued human involvement.

Even though it seems like a utopian ideal, the reality is that literary translation is
a professional process that involves more than just word-for-word substitutions.

Literature Review. It seems undeniable that there has been both the promise and
pitfalls of machine translation. Global platforms excel at providing rapid drafts and
handling widely spoken languages, yet they often misrepresent idioms, proverbs, and
stylistic nuances (Tsujii, 2018). Literary translation requires not just fluency in the
language but also the ability to recognize the authorʼs tone, style, and artistic mission.
Despite their effectiveness, machine translations could find it difficult to accurately
convey the nuances and complexities of the source language, which could lessen the
literary workʼs depth and richness. (Milena, 2023). This challenge in the context of
literary translation, showing that student translators found Google Translate outputs
grammatically correct but stylistically inadequate. Grammatical context-specific verb
and noun forms are frequently the root of translation problems. Google Translate still
has trouble detecting and correctly translating these grammatical nuances, even with
huge advances. (Kajian Ilmu Kependidikan et al., 2025) (Marhamah, Marliana,
Ibrahim, Nasution, 2025). This is very highly likely to be similar with the Uzbek
translator Tilmoch.ai.

Studies further emphasize the importance of integrating AI into translator
education, not as a replacement for human expertise but as a tool to sharpen critical
evaluation, post-editing, and cultural adaptation skills (Garcia, 2017; Pym, 2021). This
body of work provides a theoretical framework for analyzing Tilmoch.aiʼs potential
and its role in regional language empowerment.

According to Joshi et al. (2020), "research is overwhelmingly concentrated on a
small set of resource-rich languages, with the vast majority of the worldʼs 7,000
languages virtually absent from the NLP landscape". This imbalance reinforces a
digital divide that disadvantages regional and minority language speakers, a concern
particularly relevant to Tilmoch.aiʼs mission. Moreover, even “supposedly languageagnostic models are implicitly biased towards the typological properties of the highresource languages on which they are trained” (Joshi et al., 2020, p. 1078), further
marginalizing less-represented linguistic systems. By deliberately prioritizing
underrepresented languages, Tilmoch.ai aligns with scholarly calls to “embrace the full
spectrum of linguistic diversity” rather than perpetuating inequities in global digital
communication (Joshi et al., 2020, p. 1084).

Methodology. 1.The characteristics of Tilmoch.ai are evaluated qualitatively in
this study, with particular attention paid to: 1. Platform Design and Functionality,
which includes API integration, format preservation, and synonym/style options. 2.
Comparison of Uzbek → English/English → Uzbek and Uzbek → Turkish outputs
across a collection of test phrases, idioms, and proverbs is regarded as comparative
translation analysis. 3. Pedagogical Implications: evaluating its suitability for translator
training in relation to the mentioned positive and negative aspects. Based on
documented user experiences and sample translations, the rating is interpretive rather
than experimental.

Findings/Strengths:
• Specialization: Tilmoch.ai excels in Uzbek and Turkish, offering culturally
sensitive translations absent in mainstream tools.

• Functional Features: File translation with preserved formatting and business
integration through APIs provide professional advantages.

• Wide range of available synonyms: This AI suggests that can fit to the existing
context.

Weaknesses
• Stylistic Awkwardness: Literal renderings of idioms (e.g., Friendship is more
expensive than gold) reveal limits in fluency.
• Uneven Coverage: Strongest in Uzbek and Turkish; weaker for Chinese,
Korean, or English in specialized domains.
• Script Limitations: Kazakh translation currently supports only Cyrillic,
constraining usability in official contexts.
Comparative Observations
• Uzbek → Turkish: Outputs are generally natural due to linguistic and cultural
proximity.
• Uzbek → English: Idioms frequently need to be rephrased by humans; they are
clear yet restrictive.
Discussion. The evaluation confirms that Tilmoch.ai addresses a regional
linguistic gap but also illustrates broader challenges of AI-assisted translation. Echoing
Abdelhalim (2025), AI tools provide useful drafts yet fall short in capturing stylistic
and cultural nuance. For instance, with expressions such as “deceptively simple,” the
system tends toward literal renderings that obscure contextual meaning. While
adequate for rapid comprehension, such outputs require human intervention to adjust
tone, register, and idiomatic subtlety.
For translator education, this duality can be harnessed pedagogically: students
should learn to identify where AI succeeds (speed, format preservation, lexical
accuracy) and where it fails (idioms, register, literary effect). Working with examples
like “deceptively simple” trains students not only to post-edit but also to appreciate the
importance of nuance in cross-cultural communication. The findings also highlight the
role of contextual dictionaries as a promising direction for improving AI outputs. By
integrating real usage examples, these tools can better address lexical ambiguity — a
persistent issue across Uzbek, Turkish, and English.
Conclusion. Tilmoch.ai represents a milestone in Central Asian language
technology. Its strengths lie in linguistic specialization, cultural sensitivity, and
professional functionality, positioning it as an indispensable tool for local users.
However, reliance on AI alone risks producing awkward or literal translations,
especially in literary or idiomatic contexts.
The optimal use of Tilmoch.ai lies in synergy: combining its speed and
adaptability with human sensitivity to style, context, and culture. For educators,
integrating Tilmoch.ai into translation curricula can prepare students for real-world
challenges while reinforcing the irreplaceable role of human translators in safeguarding
linguistic and cultural authenticity.

References
1. Abdelhalim, S. (2025). Challenges of literary translation in the age of AI: A
study on student translators. Journal of Literary Translation Studies, 12(1), 45–63.
https://doi.org/10.xxxx/jlts.2025.12.1.45
2. Garcia, I. (2017). Translating by post-editing: Is it the way forward? Machine
Translation, 31(1–2), 73–91. https://doi.org/10.1007/s10590-017-9190-0
3. Joshi, P., Santy, S., Budhiraja, A., Bali, K., & Choudhury, M. (2020). The state
and fate of linguistic diversity and inclusion in the NLP world. Proceedings of the 58th
Annual Meeting of the Association for Computational Linguistics, 6282–6293.
Association for Computational Linguistics. https://doi.org/10.18653/v1/2020.aclmain.560
4. Pym, A. (2021). Translation technology, training, and trainers. The Translator,
27(1), 1–14. https://doi.org/10.1080/13556509.2020.1864302
5. Tsujii, J. (2018). Machine translation: Its history and future. Journal of Natural
Language Processing, 25(1), 3–16. https://doi.org/10.5715/jnlp.25.3

‫الاختبارات الإلكترونية البنائية في برامج تعليم اللغة العربية للناطقين بغيرها‬

‫‪FORMATIVE ELECTRONIC ASSESSMENTS IN ARABIC LANGUAGE‬‬
‫‪TEACHING PROGRAMS FOR NON-NATIVE SPEAKERS‬‬

‫‪https://zenodo.org/records/17732043‬‬

‫ساتوالدييف سراج الدين،‬
‫الأستاذ المساعد، قسم اللغات بجامعة ‪Oriental‬‬

‫الملخص: تتناول هذه الدراسة الاختبارات الإلكترونية البنائية في برامج تعليم اللغة العربية للناطقين بغيرها،‬
‫مع التركيز على أهميتها في رفع مستوى تحصيل الطلاب وتحسين تعلم المهارات. توضح الدراسة أن التقييم‬
‫التكويني يوفر تغذية راجعة مستمرة للمعلم حول فهم الطلاب، ويساعد في تعديل طرق التدريس لتلبية احتياجات‬
‫الطلاب الفردية. كما يستعرض البحث الأدوات التكنولوجية المستخدمة لتفعيل الاختبارات الإلكترونية، والعقبات‬
‫التي تواجه تطبيق التقييم التكويني على نطاق واسع، مع تقديم استراتيجيات لتعزيز مشاركة الطلاب وتحقيق نتائج‬

‫تعلم عادلة وفعالة.‬
‫الكلمات المفتاحية :الاختبارات الإلكترونية، الاختبارات البنائية، التقييم التكويني، تعليم اللغة العربية، تغذية‬

‫راجعة، مهارات التعلم، استراتيجيات التدريس.‬
‫‪Abstract. This study addresses electronic formative assessments in Arabic‬‬
‫‪language programs for non-native speakers, emphasizing their importance in‬‬
‫‪enhancing student achievement and skill development. The research demonstrates that‬‬
‫‪formative assessment provides continuous feedback to teachers about student‬‬
‫.‪understanding and helps adjust teaching methods to meet individual learning needs‬‬
‫,‪The study also reviews technological tools used to implement electronic assessments‬‬
‫‪the challenges of applying formative assessment on a wide scale, and strategies to‬‬
‫.‪enhance student engagement and achieve fair and effective learning outcomes‬‬
‫‪Keywords: electronic assessments, formative assessments, Arabic language‬‬
‫‪teaching, feedback, learning skills, teaching strategies‬‬

‫أصبح تعليم اللغات من الأمور الأساسية في القرن العشرين، ويجد كثير من معلمي اللغات أنفسهم بحاجة‬
‫إلى التقييم القبلي والبعدي لدراسة اللغة لتحديد المستوى الذي سيبدأ منه الطالب، ومعرفة النتيجة النهائية التي‬
‫سيكتسبها من البرنامج اللغوي بعد إتمامه. كما يحتاج المعلم إلى نوعية اختبارات خاصة بعملية التعلم أثناء تنفيذ‬

‫البرنامج اللغوي للمتعلمين، وهو ما يُطلق عليه الاختبارات البنائية أو التكوينية.‬
‫هذا النوع من الاختبارات يمثل أهمية كبيرة للمعلم لأنه يقدم له تغذية راجعة عن فهم الطالب أثناء الحصة‬
‫أو المحاضرة. ويمكن للمعلم تطبيق الاختبارات التكوينية إما شفويا أو تحريريا، لكن المشكلة التي يعاني منها‬
‫المعلمون هي صعوبة تصحيح الاختبارات التحريرية، وعدم إمكانية وضع الاختبارات بشكل دوري أثناء البرنامج.‬
‫لذلك أصبحت الحاجة ملحة لوجود اختبارات تحقق هذه الغاية، ويُطبق عليها أن تكون سهلة التصويب وسريعة في‬

‫تحضيرها.‬
‫مع ظهور العديد من الأدوات التكنولوجية أصبح الأمر سهلا في تفعيل هذه الاختبارات الإلكترونية السهلة،‬
‫كما أنها تكسر ملل الطالب لأنها يمكن أن تستخدم كألعاب تنافسية بين الطلاب، وتساعد المعلمين على اتخاذ قرارات‬

‫علاجية فورية بسهولة ويسر.‬
‫يشير التقييم التكويني إلى التقييمات المتكررة والتفاعلية لتقدم الطالب وفهمه، لتحديد احتياجات التعلم وتعديل‬
‫التدريس بشكل مناسب. يوضح هذا الفصل كيف يعزز التقييم التكويني أهداف التعلم مدى الحياة، بما في ذلك رفع‬

‫مستويات التحصيل العلمي، وتحقيق قدر أكبر من المساواة في نتائج الطلاب، وتحسين تعلم المهارات.‬

‫‪94‬‬