Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2602.14291

Bengali-Loop: Community Benchmarks for Long-Form Bangla ASR and Speaker Diarization

H. M. Shadman Tabib·Istiak Ahmmed Rifti·Abdullah Muhammed Amimul Ehsan·Somik Dasgupta·Md Zim Mim Siddiqee Sowdha·Abrar Jahin Sarker·Md. Rafiul Islam Nijamy·Tanvir Hossain·Mst. Metaly Khatun·Munzer Mahmood·Rakesh Debnath·Gourab Biswas·Asif Karim·Wahid Al Azad Navid·Masnoon Muztahid·Fuad Ahmed Udoy·Shahad Shahriar Rahman·Md. Tashdiqur Rahman Shifat·Most. Sonia Khatun·Mushfiqur Rahman·Md. Miraj Hasan·Anik Saha·Mohammad Ninad Mahmud Nobo·Soumik Bhattacharjee·Tusher Bhomik·Ahmmad Nur Swapnil·Shahriar Kabir
View original source ↗Read source document ↗Export metadata ↓

Abstract

Bengali (Bangla) remains under-resourced in long-form speech technology despite its wide use. We present Bengali-Loop, two community benchmarks to address this gap: (1) a long-form ASR corpus of 191 recordings (158.6 hours, 792k words) from 11 YouTube channels, collected via a reproducible subtitle-extraction pipeline and human-in-the-loop transcript verification; and (2) a speaker diarization corpus of 24 recordings (22 hours, 5,744 annotated segments) with fully manual speaker-turn labels in CSV format. Both benchmarks target realistic multi-speaker, long-duration content (e.g., Bangla drama/natok). We establish baselines (Tugstugi: 34.07% WER; pyannote.audio: 40.08% DER) and provide standardized evaluation protocols (WER/CER, DER), annotation rules, and data formats to support reproducible benchmarking and future model development for Bangla long-form ASR and diarization.

Explore related subjects

cs.SDeess.AS

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

H. M. Shadman Tabib, Istiak Ahmmed Rifti, Abdullah Muhammed Amimul Ehsan, Somik Dasgupta, Md Zim Mim Siddiqee Sowdha, Abrar Jahin Sarker, Md. Rafiul Islam Nijamy, Tanvir Hossain, Mst. Metaly Khatun, Munzer Mahmood, Rakesh Debnath, Gourab Biswas, Asif Karim, Wahid Al Azad Navid, Masnoon Muztahid, Fuad Ahmed Udoy, Shahad Shahriar Rahman, Md. Tashdiqur Rahman Shifat, Most. Sonia Khatun, Mushfiqur Rahman, Md. Miraj Hasan, Anik Saha, Mohammad Ninad Mahmud Nobo, Soumik Bhattacharjee, Tusher Bhomik, Ahmmad Nur Swapnil, Shahriar Kabir. 2026-02-15. Bengali-Loop: Community Benchmarks for Long-Form Bangla ASR and Speaker Diarization. https://arxiv.org/abs/2602.14291

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Xiaomi-CocktailASR-1 Technical Report

Recently, large language model (LLM) based ASR models have achieved significant progress, yet they generally lack support for multi-speaker scenarios, where the cocktail party problem remains a critical bottleneck for further advancing ASR. Existing TS-ASR methods, including end-to-end architectures with speaker embeddings and latest LLM-based explorations suffer from degraded single-speaker performance and the inability to reject when the target speaker is absent. In this paper, we propose Xiaomi-CocktailASR-1, an LLM-based end-to-end TS-ASR architecture. By utilizing reference speech as voiceprint prompts, it directly transcribes the target speaker's speech without requiring speech separation. Xiaomi-CocktailASR-1 maintains competitive performance in single-speaker scenarios, comparable to mainstream ASR models. It also features a negative sample rejection capability, outputting empty text when the target speaker is absent from the mixed speech. Additionally, Xiaomi-CocktailASR-1 supports a Chain-of-Thought (CoT) reasoning mode to provide explicit reasoning steps. Extensive experiments on various synthetic and real-world multispeaker benchmarks demonstrate that Xiaomi-CocktailASR-1 achieves state-of-the-art performance, effectively addressing the cocktail party problem through a unified architecture that balances multispeaker and single-speaker recognition accuracy, along with rejection capability.

cs.SD↗
arXiv2026

EConv-TasNet: Efficient Conv-TasNet for Effective Speech Separation

Conv-TasNet has served as a strong baseline for time-domain speech separation, and many studies have extended it with advanced architectures such as dual-path networks, U-Nets, and attention mechanisms. However, these methods often introduce high computational cost and complexity, limiting their deployment in resource-constrained scenarios. To address this issue, we propose eConv-TasNet, an efficient variant of Conv-TasNet that improves both effectiveness and efficiency without relying on resource-intensive modules. The proposed model consists of a group-wise early-splitting (GES) module and a multi-group feature aggregation (MGFA) module. GES generates discriminative speaker embeddings at intermediate stages, while MGFA progressively aggregates these group-level representations for refined mask estimation. Experimental results show that eConv-TasNet reduces model size by 22.4%, accelerates inference by 18.9%, and improves SI-SNRi by 14.0%-28.0% across three public benchmarks. Moreover, it achieves competitive performance compared with state-of-the-art methods while requiring significantly fewer parameters and lower inference cost. These results demonstrate a favorable efficiency-effectiveness trade-off for edge deployment.

cs.SD↗