[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84915-en":3,"doc-seo-84915-105":29,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},84915,1099514068035,"Ezra","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Andha-Dhun: A First Look at Audio Descriptions in Hindi","Audio Descriptions (ADs) provide narration for Blind and Low Vision (BLV) audiences during the silent gaps of movies and TV. With increasing policy momentum in India’s CBFC requirements, expanding Hindi ADs beyond English becomes essential, yet prior work for Indian languages is missing. The study presents a first systematic analysis of Hindi ADs, including the Andha-Dhun dataset of human-authored Hindi ADs from eight full-length films and two Hindi generation strategies.","arXiv :2607 .06457v 1 [ cs .CV] 7 Jul 2026  \nAndha-Dhun:  \nA First Look at Audio Descriptions in Hindi  \nRitabrata Chakraborty* 1 ,2 , Divy Kala* 1 , Nisheeth Bhooshan Gupta3 , Ganji Sreeram3 , Pailla Balakrishna Reddy3 , and Makarand Tapaswi 1  \n1 CVIT, IIIT Hyderabad, 2 Manipal University Jaipur, 3 Jio Platforms Ltd.  \n[https://github.com/katha-ai/AndhaDhun-HindiAD](https://github.com/katha-ai/AndhaDhun-HindiAD)  \n∗ denotes equal contribution  \nAbstract. Audio Descriptions (ADs) narrate visual content for Blind and Low Vision (BLV) audiences during gaps in audiovisual media. Thereis growing momentum around ADs in movies and TV shows, and with mandates from India’s Central Board of Film Certification (CBFC), there is a need to expand ADs beyond English. Yet, there is no work that generates ADs for any Indian language. To address this gap, we present the first systematic study of ADs in Hindi, contributing to aspects such as data, generation, and evaluation. We introduce Andha-Dhun, the first dataset of human-authored Hindi ADs collected from 8 full-length movies. We explore two approaches for generating ADs in Hindi: (i) directly from English dense video descriptions, and (ii) translating English ADs into Hindi. We evaluate these approaches using perplexity and LLM-as-a-judge metrics to assess fluency and quality respectively. We also analyze movies that have both English and Hindi human-authored ADs and find that naïve translation introduces artifacts and narrows diversity compared to original Hindi ADs. Direct machine translation fails to adapt cultural references, while human-translated ADs do better but still fall short. Our findings emphasize that the purpose of Hindi ADs is accessibility for Indian BLV audiences, and that this requires adapting content for the audience more than strict fidelity to the source.  \nKeywords: Audio Descriptions · Media Accessibility  \n1 Introduction  \nAudio Descriptions (ADs) are narrations inserted into silent gaps (e.g . between dialog) to help Blind and Low Vision (BLV) audiences understand visual content in movies and TV shows. An example AD is shown in Fig. 1. A sighted viewer perceives the different modes (visuals, music, speech, and sound effects) of a movie jointly to create an experience of a coherent whole. It is the job of an expert audio describer, to identify and isolate that information which is not accessible to the BLV audience, but is essential in creating that coherent whole [3] . Consequently, creating ADs requires expert knowledge and is a time-consuming and expensive process, taking up to 35-40 working hours for a 90-minute movie [14] .  \n2 R. Chakraborty, D. Kala et al.  \nHuman-authored Hindi AD  \n=रचो ने सामने हेलमेट पहने Iू टर पर आती Nई िपया को देखा ।  \nAutoAD-Zero Dense Description AutoAD-Zero AD  \nMain characters: Pia and Rancho. Actions: Pia is  \nriding a  motorcycle, while Rancho is standing and Rancho watches Pia ride a motorcycle.  \nobserving her…  \nDense-to-Hindi  \nAD-to-Hindi  \nNemotron  \nराँचो बाइक चलाती िपया को िचंितत भाव से िनहार रहा है।  \nIndicTrans2  \nरांचो िपया को मोटरसाइिकल चलाते Nए देखता है ।  \nFig. 1. Example of a Hindi audio description showing a clip from 3 Idiots with the human-authored Hindi AD. AutoAD-Zero, an automatic AD generation framework, provides a dense description of the scene in English, which is then summarized to an English AD by an LLM. We show examples of two methods to get Hindi ADs:(bottom-left) Dense-to-Hindi, obtained by using a Hindi capable LLM directly on the English dense descriptions; and (bottom right) AD-to-Hindi , where AutoAD-Zero’s [52] English AD is translated to Hindi using multiple translation models. The example shows an instance where a culture-specific item such as scooter is wrongly perceived as a motorcycle, resulting in downstream errors in Hindi generation.  \nGrowing support for ADs has led to international regulatory efforts [44], and this momentum has led to an explosion of interest in automatic AD generati","cbCaivwhaGZVXVYR","https://ap.wps.com/l/cbCaivwhaGZVXVYR","pdf",1070931,1,21,"English","en",105,"# Introduction\n## Motivation and background\n## Need for Hindi audio descriptions\n# Dataset and approaches\n## Andha-Dhun dataset\n## Dense-to-Hindi and AD-to-Hindi\n# Evaluation and findings\n## Metrics for fluency and quality\n## Translation artifacts and cultural adaptation","[{\"question\":\"What are Audio Descriptions (ADs) and who are they for?\",\"answer\":\"Audio Descriptions narrate visual content for Blind and Low Vision (BLV) audiences during silent gaps in audiovisual media, helping them understand what happens in movies and TV shows.\"},{\"question\":\"What is the Andha-Dhun dataset?\",\"answer\":\"Andha-Dhun is the first dataset of human-authored Hindi audio descriptions, collected from eight full-length movies.\"},{\"question\":\"How does the paper generate Hindi ADs and what are the main shortcomings of translation?\",\"answer\":\"It explores generating Hindi ADs either by adapting from English dense video descriptions or by translating English ADs into Hindi. Naive translation introduces artifacts, fails to adapt cultural references, and reduces diversity compared to original Hindi ADs.\"}]",1784199320,53,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"andha-dhun-a-first-look-at-audio-descriptions-in-hindi","",{"@graph":35,"@context":84},[36,53,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/andha-dhun-a-first-look-at-audio-descriptions-in-hindi/84915/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What are Audio Descriptions (ADs) and who are they for?","Question",{"text":74,"@type":75},"Audio Descriptions narrate visual content for Blind and Low Vision (BLV) audiences during silent gaps in audiovisual media, helping them understand what happens in movies and TV shows.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"What is the Andha-Dhun dataset?",{"text":79,"@type":75},"Andha-Dhun is the first dataset of human-authored Hindi audio descriptions, collected from eight full-length movies.",{"name":81,"@type":72,"acceptedAnswer":82},"How does the paper generate Hindi ADs and what are the main shortcomings of translation?",{"text":83,"@type":75},"It explores generating Hindi ADs either by adapting from English dense video descriptions or by translating English ADs into Hindi. Naive translation introduces artifacts, fails to adapt cultural references, and reduces diversity compared to original Hindi ADs.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":105,"slug":137},19,"General","general"]