Make It Hard to Hear, Easy to Learn: Long-Form Bengali ASR and Speaker Diarization via Extreme Augmentation and Perfect Alignment
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908854632054784 |
|---|---|
| author | Hasan, Sanjid Labib, Risalat Fuad, A H M Hasan, Bayazid |
| author_facet | Hasan, Sanjid Labib, Risalat Fuad, A H M Hasan, Bayazid |
| contents | Although Automatic Speech Recognition (ASR) in Bengali has seen significant progress, processing long-duration audio and performing robust speaker diarization remain critical research gaps. To address the severe scarcity of joint ASR and diarization resources for this language, we introduce Lipi-Ghor-882, a comprehensive 882-hour multi-speaker Bengali dataset. In this paper, detailing our submission to the DL Sprint 4.0 competition, we systematically evaluate various architectures and approaches for long-form Bengali speech. For ASR, we demonstrate that raw data scaling is ineffective; instead, targeted fine-tuning utilizing perfectly aligned annotations paired with synthetic acoustic degradation (noise and reverberation) emerges as the singular most effective approach. Conversely, for speaker diarization, we observed that global open-source state-of-the-art models (such as Diarizen) performed surprisingly poorly on this complex dataset. Extensive model retraining yielded negligible improvements; instead, strategic, heuristic post-processing of baseline model outputs proved to be the primary driver for increasing accuracy. Ultimately, this work outlines a highly optimized dual pipeline achieving a $\sim$0.019 Real-Time Factor (RTF), establishing a practical, empirically backed benchmark for low-resource, long-form speech processing. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_23070 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Make It Hard to Hear, Easy to Learn: Long-Form Bengali ASR and Speaker Diarization via Extreme Augmentation and Perfect Alignment Hasan, Sanjid Labib, Risalat Fuad, A H M Hasan, Bayazid Sound Artificial Intelligence Computation and Language Audio and Speech Processing Although Automatic Speech Recognition (ASR) in Bengali has seen significant progress, processing long-duration audio and performing robust speaker diarization remain critical research gaps. To address the severe scarcity of joint ASR and diarization resources for this language, we introduce Lipi-Ghor-882, a comprehensive 882-hour multi-speaker Bengali dataset. In this paper, detailing our submission to the DL Sprint 4.0 competition, we systematically evaluate various architectures and approaches for long-form Bengali speech. For ASR, we demonstrate that raw data scaling is ineffective; instead, targeted fine-tuning utilizing perfectly aligned annotations paired with synthetic acoustic degradation (noise and reverberation) emerges as the singular most effective approach. Conversely, for speaker diarization, we observed that global open-source state-of-the-art models (such as Diarizen) performed surprisingly poorly on this complex dataset. Extensive model retraining yielded negligible improvements; instead, strategic, heuristic post-processing of baseline model outputs proved to be the primary driver for increasing accuracy. Ultimately, this work outlines a highly optimized dual pipeline achieving a $\sim$0.019 Real-Time Factor (RTF), establishing a practical, empirically backed benchmark for low-resource, long-form speech processing. |
| title | Make It Hard to Hear, Easy to Learn: Long-Form Bengali ASR and Speaker Diarization via Extreme Augmentation and Perfect Alignment |
| topic | Sound Artificial Intelligence Computation and Language Audio and Speech Processing |
| url | https://arxiv.org/abs/2602.23070 |