Lipi-Ghor: A Large-Scale, Diarized, Multi-Domain Bengali Speech Corpus and a Scalable Pipeline for Building It

Bengali, spoken by over 230 million people, remains severely under-resourced in spoken language research. Existing open Bengali ASR corpora are confined to short, read or crowd-sourced speech and lack speaker diarization and long-form audio.

孟加拉语拥有超过 2.3 亿使用者，但在口语研究方面仍然严重缺乏资源。现有的开放孟加拉语自动语音识别（ASR）语料库仅限于短篇、朗读或众包语音，且缺乏说话人日志（speaker diarization）和长音频数据。

We present Lipi-Ghor, a large-scale weakly-supervised Bengali speech corpus comprising 885 hours of audio sourced from 998 YouTube videos across 596 unique channels, spanning 17 diverse domains including audiobooks, drama, religious sermons, news, and regional dialects. Each recording is annotated in SSTT format (Speaker, Speech, Transcription, Timestamp), which combines automated diarization with Bengali YouTube caption tracks.

我们推出了 Lipi-Ghor，这是一个大规模弱监督孟加拉语语音语料库，包含来自 596 个独立频道的 998 个 YouTube 视频的 885 小时音频，涵盖了有声读物、戏剧、宗教布道、新闻和地方方言等 17 个不同领域。每段录音均以 SSTT 格式（说话人、语音、转录、时间戳）进行标注，该格式结合了自动说话人日志与孟加拉语 YouTube 字幕轨道。

To our knowledge, Lipi-Ghor is the first Bengali speech corpus to pair large-scale coverage with speaker-diarized, long-form audio annotations, both of which are absent from existing open Bengali resources. We also present the construction methodology as a replicable pipeline for any language with a sufficient captioned YouTube presence.

据我们所知，Lipi-Ghor 是第一个将大规模覆盖范围与说话人日志长音频标注相结合的孟加拉语语音语料库，而这两者在现有的开放孟加拉语资源中均不存在。我们还将构建方法作为一种可复制的流程呈现，适用于任何在 YouTube 上拥有足够字幕资源的语言。

Targeted channel curation across a handful of rich domains yields corpus scales that were previously out of reach for low-resource languages. We benchmark four widely-used ASR models under zero-shot and fine-tuned conditions. All four degrade consistently on emotive, dialectal, and long-form content.

通过对少数资源丰富的领域进行有针对性的频道筛选，可以获得以往低资源语言无法企及的语料库规模。我们在零样本和微调条件下对四种广泛使用的 ASR 模型进行了基准测试。这四种模型在处理情感、方言和长篇内容时，性能均出现了一致的下降。

Because caption targets are noisy, we also release a human-annotated gold evaluation split, verified by 17 volunteer annotators over 30 hours of human-in-the-loop correction and stratified across all 17 domains. We release our original contributions (video-ID lists, diarization output, pipeline code, and the gold annotations) under CC-BY-4.0, and distribute caption-derived transcripts for research use only; no raw audio is redistributed.

由于字幕目标存在噪声，我们还发布了一个人工标注的黄金评估集，该数据集由 17 名志愿者标注员经过 30 小时的人机协同校对验证，并跨越了所有 17 个领域进行分层。我们根据 CC-BY-4.0 协议发布了我们的原创贡献（视频 ID 列表、日志输出、流程代码和黄金标注），并仅分发用于研究用途的字幕衍生转录文本；不重新分发原始音频。